[llvm] TTI: Fix special casing vectorization costs of saturating add/sub (PR #97463)

[David Green via llvm-commits]

davemgreen wrote:

I mean the old cost used to be 10, it is now 4, but something closer to 10 would be better. https://godbolt.org/z/8zs7EbzGj.

It would be nice if there was a better default than trying to cost a `llvm.sadd.sat.i64` as `2*2*cost(llvm.sadd.sat.i32)`, as that is not how these operations work.

https://github.com/llvm/llvm-project/pull/97463