> nobody has reconciled MOND with Relativity (not that it's impossible, it's just hard and annoying math, could be a lack of effort thing, could also be a real theoretical constraint that invalidates MOND).

My take on this is a bit different. The obvious way to get a relativistic theory that looks like MOND is to add a scalar field. (And possibly a vector field, as in TeVeS.) But from a relativistic, quantum field theory perspective, adding more fields is just...adding a new type of "matter". I.e., dark matter!

In other words, when you take relativity into account, "MOND" and "dark matter" don't really look like two different possibilities any more. They just look like two different ways of describing the same math in imprecise ordinary language.