[llvm] [AMDGPU] Inplace FI elimination during PEI for scalar copy instruction (PR #99556)

[Christudasan Devadasan via llvm-commits]

https://github.com/cdevadas commented:

Rewrite the PR description. Include the point that eliminateFrameIndex wasn't handling the FI copy to Scalar registers and the default implementation breaks the code while trying to handle it. There is no need to mention the two spills and the emergency spill slots.

https://github.com/llvm/llvm-project/pull/99556