[llvm] AMDGPU: Optimize mfma_scale intrinsics with 0 inputs (PR #116724)

[Matt Arsenault via llvm-commits]

arsenm wrote:

> This is failing. e.g. https://lab.llvm.org/buildbot/#/builders/127/builds/1495

Should be fixed by #117238

https://github.com/llvm/llvm-project/pull/116724