Daylight4449, something has slipped by during all this that you may be interested in discovering.
The GP40 (mentioned in the string title) was equipped with a system that Electro-Motive called "Performance Control." This system automatically adjusted the output of the locomotive to match adhesion limits compared with locomotive speed. Output was limited to 2000 HP up to approximately 12 MPH, at which point, with increasing speed, output increased until the full 3000 HP was developed at around 19 MPH. This both adapts what was, at the time, a very high hp/axle unit to low speed service and matches the locomotive's output per axle with older units to allow "mix n match" lashups. Of course, all four traction motors were operating all the time.
In point of fact, every EMD from the GP30 on that was rated higher than 500 HP/axle had this system; even the SD45 had it. So did the DD35 and DD40AX. EMD's developments led later to the Super Series controls, on the 50 series locomotives and the early Performance Control system as we know it went away.
So, you see, the question of adaptability of high horsepower units to lower speed service, as well as problems of compatibility with older lower powered units were essentially solved in the early 1960's, using a system that automatically controlled output and kept all traction motors online at all times.
I am not certain if your proposal includes the concept of reduction of prime mover / alternator output with the removal of traction motors from service, but I would point out that many locomotives DO automatically reduce output with traction motors cut out. Looking at what was closest on my desk, which is a manual for the old ALCO-GE 244-engine locomotives, I note that power output on the 1500 HP units with a pair of motors cut out was reduced automatically to 1000 HP; on the 2000 HP units, cutting out a pair automatically reduced generator output to 1300 HP. This was done through auxiliary contacts in the traction motor cutout switch which were wired into the excitation system. And this was in 1947! (For those wondering, ALCO-GE publication TP-400, pg. 819.)
One wouldn't want to cut in or out traction motors AND alter total output with increasing or decreasing speed, either; the alteration in torque on the truck frames would surely lead to instantaneous weight transfer, wheel slip and/or spin, and surging of traction current and the prime mover if it were done without reducing power completely to idle. Only one model of locomotive that I'm aware of was ever designed to do such a thing; I'm not sure if the idea reached production locomotives.
Perhaps most damaging to your idea is the fact that there are still plenty of miles of track out there that a six-axle, 420,000 lb locomotive cannot negotiate. You couldn't downgrade such a unit to run on that kind of track; it'd overturn the rail, or kink it, or break it, output corrected or motors cut out or not. You'd be getting it all dressed up with nowhere to go!
It is certainly something to think about these days that the largest road power will, in all likelihood, NOT be adaptable to downgrading for further service like the hordes of EMD GP and ALCO RS units were. However, it does seem as if your proposal is essentially an attempt to solve an already-solved problem vis a vis the mainline freight locomotives. What's really needed more than such a device is a machine designed specifically for the purpose at hand, top to bottom. Which, in many cases, is still a good old GP series unit... not ten miles from me is the old NYC/PC/CR now NS Rockport Yard, populated by deturbocharged GP50 units that are now essentially GP38's of either -2 or -3 ilk, I think. Such units did in fact have, originally, slip controls / adhesion controls designed for low speed service as described and are now more economical to run in switching, yard and transfer service without the turbochargers. They might be better described as optimized for yard service, instead of rebuilt for it.
I hope some or all of this gives you some help.