Review `if (fgGlobalMorph)` checks

[tannergooding]

As per #122368 (comment), some paths in morph restrict optimizations to only running as part of global morph and does so to avoid having to deal with VN updates. We should investigate whether these checks can be removed and whether we can keep VN up to date even if morph is called later, or whether we can stop calling morph entirely after VN.

[dotnet-policy-service]

Tagging subscribers to this area: @JulieLeeMSFT, @jakobbotsch
See info in area-owners.md if you want to be subscribed.

[MihuBot]

I'm a bot. Here is a possible related and/or duplicate issue (I may be wrong):

• JIT: Clean up fgGlobalMorph checks in morph #108875

[jakobbotsch]

Yes, this was already done with #108875. The remaining checks are not for the sake of CSE candidates.

[EgorBo]

I think most of the if (fgGlobalMorph) still should be removed from morph. It's just that today that will most likely lead to CQ regressions when called from late phases, because morph erases VN during transformations. It should update it instead.