[InstCombine] reduce vector casting before icmp

There may be some generalizations (see test comments) of these patterns,
but this should handle the cases motivated by:
https://llvm.org/PR51315
https://llvm.org/PR51259

The backend may want to transform differently, but at least for
the x86 examples that I looked at, there does not appear to be
any significant perf diff either way.