[llvm] [AArch64] Use SVE for load-extend-broadcast on V3AE (PR #221689)

[Sander de Smalen via llvm-commits]

================
@@ -1022,6 +1022,10 @@ def FeatureLimited64bitVectorMulBandwidth : SubtargetFeature<
     "Has limited 64bit vector multiply bandwidth compared to scalar multiply",
     [], InlineIgnore>;
 
+def FeaturePreferSVEVectors : SubtargetFeature <
----------------
sdesmalen-arm wrote:

There already seems to be a `FeatureNoSVEFPLD1R` feature, which seems rather similar. Can this feature be reused for this transform?

https://github.com/llvm/llvm-project/pull/221689