[llvm] [X86][LoopVectorize] Enable MaximizeBandwidth by default on X86 with generic split cost floor (PR #201666)

[Sumukh J Bharadwaj via llvm-commits]

amd-subharad wrote:

> It would probably be good to share some data on the performance impact on a few different X86 systems. IIRC there were some regressions on other targets when enabling this

Sure we do have results showing improvements from better VF selection. Will share the detailed numbers

https://github.com/llvm/llvm-project/pull/201666