The bigger issue is size. A train can brake at each car, not just the locamotive. So that means hundreds of regenerating electric motors spread across the train. Dump that all to a collective "third rail" and all manner of electrical havoc will begin as out-of-phaze motors start driving each other to do the wrong thing.
Instead of a freight train lets consider the slightly aged (god I'm old) electric train I would once have caught when I worked in London. As you say, each set of wheels can be braked independently and they're all using regenerative electric motors, so that's maybe 40 wheel sets on the train I'm thinking of. But, traction current is supplied at only a maximum of four points on that train, I'd guess only one is actually ever in use for simplicity.
So instead of "electrical havoc" it's a pretty simple local problem for the engineers designing the train.
For freight trains the power is applied at the locomotive only. However for passenger trains it is very common for each car to have drive wheels. This allows for more acceleration and thus more traction. There is no technical reason you can't do this for a freight train, but it is some engineering work that isn't normally done. It is a little tricky for this application because typically ore is loaded through the top of the car and unloaded underneath which makes finding a place to get the power to the motors tricky, and dust from loading/unloading can wreck things - these are solvable problems but still problems.
Conventional freight wagons are unpowered yes, but I don't see how that's relevant to this claim that somehow if you have multiple electric motors you can't dump power into the traction feed when decelerating. British passenger trains do this today, it's not even a quirk, it's the default. The local train operator boasts that they dump 78,700,000 kWh per year from deceleration back into the traction feed. [No I don't know why they picked those units]
