[AMDGPU] Use WriteSALUDummy for v_mad_u64_u32 (#191500)
Updates the the new Write*Dummy HWWriteRes to maintain latency -- this
is to ensure SchedModel->computeOperandLatency calls are unchanged.
This also uses the new HWWriteRes for v_mad_u64_u32 and v_mad_i64_i32.
For an explanation of why we want to do this , see
https://github.com/llvm/llvm-project/pull/190095 . In short, the
scheduler will not try to cover the full latency of the instructions
without this new modelling.
This causes a good amount of lit churn since these instructions are well
covered.
