[llvm] [VPlan][LoopVectorize] Truncate min/max intrinsic ops (PR #90643)

[Patrick O'Neill via llvm-commits]

patrick-rivos wrote:

Rebased to resolve conflicts with e846778e52f8586b5092c2fd4cdbec2334e31770.

Fixup to handle new VPWidenCall arg and clang-format.

I'm a bit wary of the false-dependency bits for bswap intrinsics to prevent `i1` bswaps, but I haven't been able to come up with a testcase where it negatively impacts bit-minimization.

Also realized I forgot to ask for review: VPlan so @fhahn when you have the time could you take a look at this?

https://github.com/llvm/llvm-project/pull/90643