[llvm] [VPlan] First step towards VPlan cost modeling. (PR #92555)

[via llvm-commits]

jeanPerier wrote:

FYI the new assert `VF.Width == Width && "VPlan cost model and legacy cost model disagreed"` is also firing when compiling Fortran benchmark SPEC ACCEL calculix and cam. These applications are huge, so I will need time to give a reproducer/insructions.
I think my tests included your https://github.com/llvm/llvm-project/commit/eea150c84053035163f307b46549a2997a343ce9 fix, but I also need to double check that. I will come back to you next week.

https://github.com/llvm/llvm-project/pull/92555