For the Allison it is a question of what was actually available when.
The V-1710-45 (F7) was a 1940 project using the two stage system. The contract was dated Dec 2nd 1940, for one experimental engine.
However change order #1 was issued in June of 1942 and included changing the impeller from 9.5 in to 12.1875 in and changing from the friction clutch to the hydraulic clutch.
The original delivery date of Sept 1941 was changed to February 1943 so one can see how badly the program was slipping (free fall?)
A 2nd contract had been signed in Dec of 1941 for an experimental engine with intercooler.
It also does not appear that you can simply slap a 2nd stage on an engine, fit an "acme"* intercooler and "Bob's your uncle" have an engine that equals a two stage Merlin.
Unless I am reading it wrong This early Allison two stage was fitted with a liquid cooled after cooler, weighed 1545lbs and was to be rated at 1150hp military at 21,000ft running on 100/125 fuel.
Compare that to a Merlin 46 single stage engine.
The Allison V-1710-101 engine was first tested in July 1943 but performance was still a little behind. 1500hp WER at 6000ft (running at 3200rpm) wasn't going to set the world on fire and 1150hp at 20,000ft at 3,000rpm wasn't going to cause anything else to be canceled. There were some modifications done.
The Allison V-1710-121 is the one that matches the Merlin 14SM but it isn't even tested until early 1944.
1700hp at 26,000ft (3200rpm) with water injection is very good. It is too late (and it took quite a while for Allison to sort out the -121 engine, if it was ever fully sorted out?)
The USAF "Reciprocating Engine Characteristics" chart lists the -45 as 1515 lbs, but your source may be more accurate (in any case that's little more than the Merlin 20 series single stage 2-speed drive or Packard V-1650-1 at 1520 lbs). If the earlier simple friction clutch was lighter that would've put it more as single-stage 2-speed Merlin competition, but the greater length would still make it more difficult to install on existing production designs. I'm assuming you mean simple mechanical clutch with friction plate for synchronous gear shifting for the "friction clutch" and a fluid coupling torque converter impeller + turbine for the hydraulic one (and not just hydraulic actuation of a mechanical clutch).
If the early mechanical clutch design was still only single-speed, you'd have low blower when de-clutched (with aux-stage free-wheeling similar to when the fluid coupling has no oil in it) and high blower with it engaged. In that context, the intercooler would be more useful/necessary (even conventional air-to-air ones) due to the charge heating you'd get with the aux stage engaged. OTOH, if the integral stage was already geared 8.8 for medium altitude (FTH reduced somewhat due to drag through the pinwheeling aux stage) and single-speed aux stage providing FTH similar to the V-1650-1 or optionally higher than that (more like the Merlin 47) you'd still have low-blower for reasonable take-off and low altitude performance. (the drag from the aux stage should actually allow more take-off power at given manifold pressure than normal 8.8 supercharged engines due to that reduced FTH meaning less drag from the throttle plate and smoother drag from the "dead" aux impeller acting like a swirl throttle)
The big problem with the aux stage V-1710 isn't even it's timing, but inability to be bolted-on to existing designs with minimal modification. If they'd used a bevel gear drive to run the aux stage perpendicular to the integral stage, flush with the port or starboard side of the engine (like the DB or Jumo engines) it should've avoided that issue. Even if the installation protruded and added to frontal area, it should've been better, and the location of a ram intake on the side would've improve the P-40 and P-51's view over the nose slightly, while may have actually improved RAM pressure on the P-39. If they could've done that sort of arrangement with an earlier, more primitive single-speed aux drive, it could've been worthwhile. (2-speed drive on the aux stage would be better, too)
Now the actual production aux stage V-1710s had the advantage of that fluid coupling and smooth power curve along with reduced weight and complexity over an intercooler arrangement, but with the trade-off of charge heating at higher altitude (like the early P-38s or like using a turbo without intercooler, but probably not as bad as a turbo: either due to efficiency of compressor design or additional heating from the exhaust/turbine section, which is very closely coupled on GE turbos at least). However, once the P-51 was in production, it'd have been smarter to contract out modifications for a 2-stage 1710 version while NAA was occupied with Merlin Mustang production, but that's assuming good cooperation and interest between licensees (and from Curtiss's end you could consider other things like licensed F4U production or P-38 production: particularly handing off development of a non-turbocharged P-38 variant to Curtiss with proper ejector exhausts and ram air intake, both with the single and 2-stage alisons). Start with the P-51A as the basis, or start before that with a license given for the initial P-51 design and developing that further in parallel with NAA's efforts. You could simplify some things, omit the bubble canopy, maybe introduce P-40 style scaloping for better rear view along with a sliding canopy with blown out malcolm hood style for improved production (if adapting the P-40's canopy itself was feasible for initial production, transition to a hood would've been much easier to field-mod than the hooded Mustangs, too).
There is one other area specific to Curtiss that could have benefited the V-1710: cooperation with Curtiss-Wright's engine division. That could range from actual consult work to assist in supercharger development, or simply technology sharing: in the 1938/39 timeline range, you have both the R-1820 and R-2600 with larger superchargers and 2-speed drives that might have made a reasonable basis for an auxiliary supercharger on the V-1710. And, aside from using them as a second stage, the integral stage could also be deleted in favor of just a single 2-speed auxiliary stage without the same space constraints of the integral supercharger's location. (that deletion might also allow correcting the kink/bottleneck in the V-1710's induction system, or at least improving it somewhat: I forget if the kink was at the carb/supercharger inlet or downstream of the supercharger outlet, but I think the former, which should mean re-arranging the intake should help). A short-cut for that deletion might have been fitting a dummy impeller disc (if flywheel stability mass was needed), using the supercharger housing as a simple radial intake duct fed from one of the sides, and extending the impeller shaft outside of that casing so it could be used as the external supercharger drive (potentially with a bevel gear), that-is, if it was easier to do that than using the existing accessory drive gears to power the external supercharger. An R-2600 supercharger + V-1710 would also probably be more useful than any R-2600 design studies on the P-36/P-40 airframe.
Overall better technology-sharing between firms during war-time (and potentially waiving license/royaltee fees during war-time) between more engine manufacturers on both sides of the Atlantic (like RR sharing with Allison for that matter, or licensing the Merlin XX supercharger as the Allison's aux stage) may have helped a lot, but even aside from that, the Curtiss-Wright connection would've allowed potential cooperation well before the US entered the war and potentially before the war started in '39. (experiments with Wright superchargers on Allison engines starting in '38/39 resulting in small scale production engines by '41and full scale by '42 might have helped a lot) There could've been contractual conditions with Wright pre-war that gave them exclusive use of those engine models, but may have been later pressured into waiving that some time in 1940-41. (or something more like Wright either supplying the superchargers to Allison or getting royalties for each unit built: given Wright's problematic quality control during war-time, having Allison build them would've been better ... though pre-war or early war that may not have been a problem and I don't think superchargers were generally the point of failure with Wright's production, R-2600 or otherwise) But if it could've been arranged, sharing P&W's aux stage supercharging developments with Allison may have been more helpful (both from the R-1830 and R-2800).
Allison also probably would've had more resources to work on more V-1710 features (supercharger developments, water injection for super or turbo installations, different compression ratios) has the US not been investing in so many experimental, early development stage designs, like the Army's entire Hyper engine project and had undergone some sort of "rationalization" scheme for war-time like Britain did, aside from turbojet developments at least. (albeit prior to the Whittle design being brought over, I think all US turbojet developments were internally/privately funded efforts, and mostly or entirely limited to "paper" design studies)
Given GE's slower pace at improving impeller/diffusor efficiencies early-war, some sort of technology sharing on that front should've been helpful as well. (among other things, it might have even made some intercooler-less turbo installations viable, particularly if there was better compressor + turbine separation/isolation/insulation, or longer shaft arrangements designed specifically to nest behind a V-12 engine with the turbine shaft horizontal and perpendicular to the engine's drive shaft, with the compressor on one side and turbine section on the other, and the turbine shaft nested up against the crankcase) That or GE could have just supplied turbine drive sections a la carte and allowed engine companies to design and mount their own compressor sections to. (I think Wright did some of their own turbocharger development during the war and one or more was tested in the XP-60 project, but having the "hot" section ready-made from GE, offloading that end of development, might have allowed the best of both worlds)
But back onto the P-40 itself, I think it's fair to compare it directly to the Spitfire and 109 in design quality and potential for growth (or the base P-36 to the Spitfire Mk.I and 109D and E). The P-36 was larger and had much larger fuel capacity from the start, and maintained more internal fuel tankage even as it decreased up to the D/E with 149 US gal (then increased slightly again with the F/K to 157, then down to 122 in the stripper P-40N, but 161 gal with the non-stripper N versions). And comparing it to 2-stage merlin, let alone Gryphon engined Spitfires is hardly fair as US production didn't have those available (or had production consumed by the P-51B/C/D already). OTOH you could argue the lower weight of the 2-stage allison might benefit the Spitfire more than it did any P-40 or P-51 installations. (though it should've helped with range and/or specific fuel consumption: ie same range on less fuel in a hypothetical war-time Allison P-51) The P-40 is somewhat bigger, somewhat heavier, but pretty clean/low drag considering that vs the 109 and Spitfire, it had much better high speed control forces and handling from the start (and the Spitfire played catch up) and higher rate of roll even at low speeds, giving it turn-reversal advantages in combat, plus faster split-S diving (and better dive acceleration due to sheer weight + clean airframe) to allow for diving as an escape maneuver when wishing to leave combat.
Also note: part of why the P-40 was heavier was the same reason all US fighters tended to be overweight (even when unarmored pre-war), and that was due to higher max G-force specifications from the USAAC (and I believe USN as well, though carrier fighters need added strength for landing, too).