Land forces are much more flexible and adaptable (and considerably cheaper) than the air or sea services. You can take an armored battalion or an air defense battalion, park the vehicles somewhere, and use the personnel for a variety of purposes; we've been doing just that for years. When I was a tank battalion S-3 we guarded Haitian refugees at Gitmo and fought prairie fires - and this was in 1995. So you can design an army for high-intensity warfare and still be comfortable that you will have at least some capacity for lesser contingencies.
People are always going to be more adaptable than equipment - that much is obvious. So too is the fact that air and naval forces are more dependent on equipment to operate in their domains and so can be considered less flexible. But for "high tech" warfare, the Army is pretty much just as equipment dependent as the Navy and Air Force.

Force structure would be even more different. What is not commonly realized is that cost is less of a consideration in force structure than manpower. There is always more money available, but congressionally mandated manpower caps are much more difficult to move. So the 10,000 or so bodies that we invest in a carrier battle group (a swag from a groundpounder, by the way, so don't quote me) would ideally be reinvested in brown water forces, CBs, special forces, or in the Army.
You're completely correct about the limitations of manpower on force structure, but I think you're completely wrong in the assertion that manpower is cheap. It is not, especially once one considers that some legacy manpower costs are not part of the defense budget. Moreover, the budget for personnel is much less flexible. You can't save money and divert it to other things nearly as easily as you can with procurement and O&M money.

Additionally, moving personnel from one area (carrier battle group) to another (brown water, CB, SF, Army) is neither easy nor cheap. The skills and equipment requirements are quite different and, in a volunteer force at least, one cannot simply order that nuclear reactor tech to become a SEAL or CB or whatever. You can force them out and recruit replacements, or offer incentives to change, but both of those are expensive, and that's not even discussing the recruiting, retraining, decommissioning costs as well as costs to equip the force for the new task. And by "costs" I'm talking both money AND time. Changing force structure is therefore an expensive and slow process.

Not so much for the really expensive parts of the armed forces. Yes, carrier battle groups and F-16 squadrons have utility in small wars - but if you are designing a Navy or an Air Force to support wars like Afghanistan or Iraq, and 'accepting risk' like Sec Gates says we are, the weapon systems you buy would look much different from the ones we are currently acquiring. You would want air frames, for example, designed almost solely to accurately deliver ordnance (or bags of food); air superiority would not be a consideration.
Turn your argument on its head. Suppose we build a specialized force for small wars - what happens when a big war comes around? Then you're stuck with capabilities you can't use and then the argument is turned upside down. Better, I think, to have capabilities that can do both imperfectly than try to bet the farm on what the next war is going to be and create an ideal force for that particular war. This is actually what both the Air Force and Navy have been doing for almost 20 years now - getting rid of "one-trick-pony" capabilities in favor of more flexible capabilities. Does that have costs? Sure, but what is the alternative? Try to reconfigure most of your force with every new conflict? One can certainly do that to some extent (and all the services are), but changing a carrier battle group into something optimized for small wars may not be practical or wise for a whole host of reasons.