Fix a bug in `FlowGraphNaturalLoop::MatchLimit`; it did not properly check limit
locals for address exposure when evaluating their invariance (the old logic does
that). Also support promoted struct fields (like the old logic).

No diffs expected -- the diffs of the quirk here were coming from this bug in
`FlowGraphNaturalLoop::MatchLimit`.

With the new loop representation there will never be any `BBJ_RETURN`
blocks in loops we are cloning -- add an assert for this. Then remove
the quirk that avoids cloning loops when the old loop cloning would
create too many `BBJ_RETURN` blocks.

Large size wise regressions are expected due to new cloning.

…ng (#96504)

* Assembly.GetManifestResourceNames, GetReferencedAssemblies

* AssemblyLoadContext.GetLoadedAssemblies

* RuntimeModule.GetTypes

* Delegate.BindToMethodName

* Delegate.BindToMethodInfo

* Delegate.GetCallStub, Delegate.AdjustTarget

* Delegate.InternalAlloc, InternalAllocAlike

* Delegate,FindMethodHandle, InternalEqualMethodHandles

* GC.GetTotalAllocatedBytes

* GC.WaitForFullGCApproach, WaitForFullGCComplete

* Update src/coreclr/vm/commodule.cpp

Co-authored-by: Aaron Robinson <[email protected]>

* Apply suggestions from code review

* CR feedback, cleanup

- Delete dead code
- Move data first in RuntimeModule.cs

---------

Co-authored-by: Aaron Robinson <[email protected]>

…#96584)

* Fix CA1820

* Fix IDE0003

* Fix IDE0004

* Fix IDE0005

* Fix IDE0045

* Fix IDE0038

* Fix IDE0051, IDE0052

* Fix IDE0055

* Fix IDE0066

* Fix IDE0078

* Fix IDE0290

* Fix IDE0300

* Fix IDE0301

* Fix IDE0270

* Fix IDE0079

Fixed 4 links to other pages in this repo.

* Fix FrozenObjectHeapManager compilation for 32-bit targets

* Fix DehydratedDataNode throwing exceptions on 32-bit platforms

* Fix llvm-libunwind compilation for EHABI targets

* Fix up compilation of bit-rotten NativeAOT ARM code