[llvm] [AMDGPU][True16][CodeGen] add a 16bit d16 predicate for true16 mode (PR #156574)

[Joe Nash via llvm-commits]

https://github.com/Sisyph commented:

This part of the patch looks good. But some more handling is needed for spills. For example, spillv16.ll still has scratch_load_d16_b16

https://github.com/llvm/llvm-project/pull/156574