[llvm] [AArch64] Fix heuristics for folding "lsl" into load/store ops. (PR #86894)

[David Green via llvm-commits]

================
@@ -1268,7 +1270,6 @@ def TuneNeoverseN1 : SubtargetFeature<"neoversen1", "ARMProcFamily", "NeoverseN1
                                       "Neoverse N1 ARM processors", [
                                       FeatureFuseAES,
                                       FeatureFuseAdrpAdd,
-                                      FeatureAddrLSLFast,
----------------
davemgreen wrote:

+FeatureAddrLSLSlow14 here I would expect?

https://github.com/llvm/llvm-project/pull/86894