[llvm] Use (de)interleaving intrinsics for fixed-length vectors (PR #202583)

[Kamlesh Kumar via llvm-commits]

kamleshbhalui wrote:

> The AArch64 neon improvements look good, I don't see any regressions.
> 
> Would you mind enabling this by default for all targets just so we can see what the impact is on e.g. RISC-V + X86?

I have enabled for all, we could get the result for RISCV as well now.
Regarding X86, It does not support lowering of these intrinsics.

https://github.com/llvm/llvm-project/pull/202583