[llvm] [SelectionDAG][AArch64] Add dot product lowering in NEON for PARTIAL_REDUCE_*MLA ISD nodes (PR #140075)

[Benjamin Maxwell via llvm-commits]

================
@@ -29518,37 +29538,64 @@ SDValue AArch64TargetLowering::LowerVECTOR_HISTOGRAM(SDValue Op,
 }
 
 /// If a PARTIAL_REDUCE_MLA node comes in with an accumulator-input type pairing
-/// of nxv2i64/nxv16i8, we cannot directly lower it to a (u|s)dot. We can
+/// of v2i64/v16i8, we cannot directly lower it to a (u|s)dot. We can
----------------
MacDue wrote:

Nit: Maybe but "nx" in brackets e.g.` (nx)v2i64`, since this applies to both scalable and fixed vectors. 

https://github.com/llvm/llvm-project/pull/140075