[llvm] [VectorCombine] Prevent redundant cost computation for repeated operand pairs in foldShuffleOfIntrinsics (PR #171965)

[Simon Pilgrim via llvm-commits]

https://github.com/RKSimon requested changes to this pull request.

Next step is to confirm it works for #170867 and add the test to llvm\test\Transforms\VectorCombine\X86\shuffle-of-intrinsics.ll

https://github.com/llvm/llvm-project/pull/171965