[llvm] [InstCombine] Fold uitofp of masked truncations (PR #222592)

[Harrison Hao via llvm-commits]

harrisonGPU wrote:

> LGTM. I don't have a better idea for generalizing this pattern. Perhaps we can use `canEvaluateZExtd + EvaluateInDifferentType`. But it needs a hint for the wider bit width.

Thanks for the helpful suggestion. I’ll explore it after I finish this PR.

https://github.com/llvm/llvm-project/pull/222592