[llvm] [X86] Try Folding icmp of v8i32 -> fcmp of v8f32 on AVX (PR #82290)

[via llvm-commits]

goldsteinn wrote:

> Reposting here. If the input bound checks fail, `vpcmpeqd` can be emulated with no restrictions on the input bounds:
> 
> ```
> vpcmpeqd result, a, b
> =>
> vpxor ymm0, a, b
> vcvtdq2ps ymm1, ymm0
> vcmpps result, ymm1, zero, 0
> ```

Ah nice :)

Ill add todo for that.

https://github.com/llvm/llvm-project/pull/82290