[llvm] [InstCombine] Transform high latency, dependent FSQRT/FDIV into FMUL (PR #87474)

[Joshua Cranmer via llvm-commits]

https://github.com/jcranmer-intel commented:

I haven't gone through all of the thinking on fast-math flags yet, but I've noted at least one incorrect flag:

https://github.com/llvm/llvm-project/pull/87474