[llvm] [AMDGPU][DAGCombiner][GlobalISel] Prevent FMA contraction when multiply cannot be eliminated (PR #169735)

[Adel Ejjeh via llvm-commits]

adelejjeh wrote:

After discussing with @b-sumner and seeing that some lit tests for non-AMDGPU targets are failing, we agreed it would be best to guard the change to only apply for AMDGPU target. I made that change.

https://github.com/llvm/llvm-project/pull/169735