[X86][Costmodel] getReplicationShuffleCost(): promote 1 bit-wide el…

…ements to 32 bit when have AVX512DQ

I believe, this effectively completes `X86TTIImpl::getReplicationShuffleCost()`
for AVX512, other than the question of handling plain AVX512F,
where we end up with some really ugly "shuffles",
but then is there any CPU's that support AVX512, but not AVX512DQ/AVX512BW?

Reviewed By: RKSimon

Differential Revision: https://reviews.llvm.org/D114315

llvm/lib/Target/X86/X86TargetTransformInfo.cpp

@@ -3772,6 +3772,10 @@ X86TTIImpl::getReplicationShuffleCost(Type *EltTy, int ReplicationFactor,
         PromEltTyBits = 16; // promote to i16, AVX512BW.
       break;
     }
+    if (ST->hasDQI()) {
+      PromEltTyBits = 32; // promote to i32, AVX512F.
+      break;
+    }
     return bailout();
   default:
     return bailout();