This is a performance optimisation for subtyping between two unions thatare largely the same.Fixes#14034This makes@adriangb's example in#14034 (comment)finish basically instantly. I could add it as a unit test?Type checking pydantic core is still not fast — takes like four or fiveminutes with uncompiled mypy — but at least it's now feasible. I thinkthere's room for doing some optimisation in make_simplified_union thatwould improve this.

The following code optimises make_simplified_union in the common casethat there are exact duplicates in the union. In this regard, this issimilar topython#15104To get this to work, I needed to use partial tuple fallbacks in a coupleplaces (these maybe had the potential to be latent crashes anyway?)There were some interesting things going on with recursive type aliasesand type state assumptionsThis is about a 25% speedup on the pydantic codebase and about a 2%speedup on self check (measured with uncompiled mypy)

Fixes#15192The following code optimises make_simplified_union in the common casethat there are exact duplicates in the union. In this regard, this issimilar to#15104There's a behaviour change in one unit test. I think it's good? We'llsee what mypy_primer has to say.To get this to work, I needed to use partial tuple fallbacks in a coupleplaces. These could cause crashes anyway.There were some interesting things going on with recursive type aliasesand type state assumptionsThis is about a 25% speedup on the pydantic codebase and about a 2%speedup on self check (measured with uncompiled mypy)