[AMDGPU] Reimplement the GFX11 early release VGPRs optimization

Implement this optimization in SIInsertWaitcnts, where we already have
information about whether there might be outstanding VMEM store
instructions. This has the following advantages:
- Correctly handles atomics-with-return.
- Correctly handles call instructions.
- Should be faster because it does not require running a separate pass.

Differential Revision: https://reviews.llvm.org/D153279