[llvm] [AMDGPU] Added hot-block-rematerialize pass (PR #126331)

[David Stuttard via llvm-commits]

dstutt wrote:

We've got some work in the pipeline to allow pre-RA machinescheduler to run in SSA form as well, so that might be ok.
Also, this might work well with some other work we're considering to add spilling as a separate pass before RA (but in SSA form).

https://github.com/llvm/llvm-project/pull/126331