[llvm] [AMDGPU] Introduce Next-Use Analysis for SSA-based Register Allocation (PR #156079)

[Nicolai Hähnle via llvm-commits]

https://github.com/nhaehnle requested changes to this pull request.

There is *way* too much code in header files.

Frankly, this smells like a prime example of an AI-generated PR that is being handled badly.

https://github.com/llvm/llvm-project/pull/156079