[llvm] [LLVM][CodeGen][SVE] Clean up lowering of VECTOR_SPLICE operations. (PR #91330)

[Paul Walker via llvm-commits]

paulwalker-arm wrote:

This loosely relates to https://github.com/llvm/llvm-project/pull/90963 whereby tightening the lowering of VECTOR_SPLICE means we can use all variants when lowering other operations.  Whilst this PR does not fix a bug within current code it will prevent a situation whereby a BITCAST can be created with illegal types post type legalisation.

https://github.com/llvm/llvm-project/pull/91330