Revert earlier change that flagged ds permute operations as not affecting lgkm counter.

[kanarayan]

Summary

https://bugs.llvm.org/show_bug.cgi?id=32666 found a case with memory operands. Need to study this better and correct the new pass.

[arsenm]

LGTM but a test would be good

[kanarayan]

I will add the test from the bug report as a lit test separately.

[arsenm]

In D32254#731305, @kanarayan wrote:

I will add the test from the bug report as a lit test separately.

All that is probably required is that there is a waitcnt lgkmcnt in the existing test for the bpermute intrinsic

lib/Target/AMDGPU/DSInstructions.td

 class DS_1A1D_PERMUTE <string opName, SDPatternOperator node = null_frag>
 : DS_Pseudo<opName,
   (outs VGPR_32:$vdst),
   (ins VGPR_32:$addr, VGPR_32:$data0, offset:$offset),
   "$vdst, $addr, $data0$offset",
   [(set i32:$vdst,
    (node (DS1Addr1Offset i32:$addr, i16:$offset), i32:$data0))] > {
 
-  let LGKM_CNT = 0;
-
   let mayLoad = 0;
   let mayStore = 0;
   let isConvergent = 1;
 
   let has_data1 = 0;
   let has_gds = 0;
 }