[llvm] [AMDGPU] Add IR LiveReg type-based optimization (PR #66838)

[Matt Arsenault via llvm-commits]

arsenm wrote:

> It is mainly due to this special condition entry that the hook will not work for the feature in this PR. Not only that special condition, but the generic codegenprepare optimizePhiType is disabled for vector types. Of course we could remove the conditions to the target hook, but this feels like misusing the hook since the features are conceptually different and we will have a phase ordering problem (code sinking will fold the newly inserted casts).

The implementation is still a lot simpler, if the heuristic and placement is different. Does copying the same technique, and using the existing ValueToValueMap work? 

https://github.com/llvm/llvm-project/pull/66838