[llvm] [LV][AArch64] LoopVectorizer allows scalable frem instructions (PR #76247)

[Paschalis Mpeis via llvm-commits]

paschalis-mpeis wrote:

> We are already checking if a library function exists, wouldn't it be more accurate to replace to widen recipe with a WidenCall recipe with the function?

@fhahn unfortunately, I will need a `CallInst` to do that, as `tryToWidenCall` / `VPWidenCallRecipe` and their dependencies do not seem to be easily adaptable for `Instructions`.

https://github.com/llvm/llvm-project/pull/76247