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

[Timothy Herchen via llvm-commits]

anematode wrote:

Reposting here. If the input bound checks fail, `vpcmpeqd` can be emulated with no restrictions on the input bounds:

```
vpxor xmm0, a, b
vcvtdq2ps xmm1, xmm0
vcmpps xmm1, zero, 0
```


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