[llvm] [AArch64][GlobalISel] Adopt some Ld* patterns to reduce codegen regressions (PR #135492)

[David Green via llvm-commits]

================
@@ -573,3 +573,10 @@ def : Pat<(v2i64 (AArch64dup (i64 (load GPR64sp:$Rn)))),
           (LD1Rv2d GPR64sp:$Rn)>;
 def : Pat<(v1i64 (AArch64dup (i64 (load GPR64sp:$Rn)))),
           (LD1Rv1d GPR64sp:$Rn)>;
+
+defm : LoadInsertPatterns<load, v16i8, v8i8, nxv16i8, i8,
----------------
davemgreen wrote:

If we do go this route I would put these next to the existing LoadInsertPatterns, maybe with a comment about it being used for GISel. Same for the others below.

https://github.com/llvm/llvm-project/pull/135492