[llvm] [AMDGPU][CodeGen] Implement SimplifyDemandedBitsForTargetNode for readfirstlane. (PR #190009)

[via llvm-commits]

lakreite wrote:

> > Propagate demanded bits through readfirstlane and readlane intrinsics in AMDGPUISelLowering with SimplifyDemandedBitsForTargetNode implementation.
> 
> Update the description: the patch does not handle readlane.

Description is updated. Thank you.

https://github.com/llvm/llvm-project/pull/190009