Remove ax-riotaBAD [mathbox]

[icecream17]

https://us.metamath.org/mpeuni/ax-riotaBAD.html

As noted in #4248 there are several parts of set.mm with bad theorems, so this issue is about one "family" of them: those using ax-riotaBAD (~2.8% of set.mm)

This axiom is only used in https://us.metamath.org/mpeuni/riotaclbgBAD.html, where the inconsistency with set.mm is in the reverse direction. I'd expect that most of the time, deeper in the dependency chain (for instance https://us.metamath.org/mpeuni/glbconN.html), it should be possible to just prove uniqueness directly.