[AMDGPU] Optimize VGPR LiveRange in waterfall loops

The loops are run exactly once per lane, so VGPRs do not need to be
saved. Use the SIOptimizeVGPRLiveRange pass to add phi nodes that take
undef when coming from the loop.

There is still a shortcoming:
Return values from a function call in the loop are copied because their
live range conflicts with the live range of arguments, even if arguments
are only IMPLICIT_DEF after the phi insertion.

Differential Revision: https://reviews.llvm.org/D105192