Collapse VM contracts to STANDARD_VM_CONTRACT where equivalent

[AaronRobinsonMSFT]

Collapse expanded CONTRACTL blocks to STANDARD_VM_CONTRACT / STANDARD_VM_CHECK in the VM where semantically equivalent.

Helpers reachable from both cooperative and preemptive callers (metadata capture, static-field-address) are kept mode-agnostic to avoid adding an incorrect MODE_PREEMPTIVE check.

Validated with a Checked x64 build and the full System.Reflection test suite (1794 passed, 0 failed).

Note

This PR description was generated with assistance from GitHub Copilot.

[azure-pipelines]

Azure Pipelines:
Successfully started running 3 pipeline(s).
13 pipeline(s) were filtered out due to trigger conditions.
There may be pipelines that require an authorized user to comment /azp run to run.

[dotnet-policy-service]

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

[jkotas]

Looks like we have a problem with sprintf in CONTRACT_ASSERT macro:

000000d1`1e1f8dc0 00007fff`4e072c97 ntdll!KiUserExceptionDispatcher+0x3a
000000d1`1e1f9b30 00007fff`4e073370 coreclr!common_strnlen_c<0,unsigned char>+0x57
000000d1`1e1f9b70 00007fff`4e072b26 coreclr!common_strnlen_simd<0,1,unsigned char>+0xc0
000000d1`1e1f9d00 00007fff`4e073f4d coreclr!common_strnlen<0,unsigned char>+0x26
000000d1`1e1f9d30 00007fff`4e069458 coreclr!strnlen+0x1d
000000d1`1e1f9d60 00007fff`4e068f62 coreclr!__crt_stdio_output::output_processor<char,__crt_stdio_output::string_output_adapter<char>,__crt_stdio_output::format_validation_base<char,__crt_stdio_output::string_output_adapter<char> > >::type_case_s_compute_narrow_string_length+0x28
000000d1`1e1f9d90 00007fff`4e062f52 coreclr!__crt_stdio_output::output_processor<char,__crt_stdio_output::string_output_adapter<char>,__crt_stdio_output::format_validation_base<char,__crt_stdio_output::string_output_adapter<char> > >::type_case_s+0x102
000000d1`1e1f9dd0 00007fff`4e05e5cf coreclr!__crt_stdio_output::output_processor<char,__crt_stdio_output::string_output_adapter<char>,__crt_stdio_output::format_validation_base<char,__crt_stdio_output::string_output_adapter<char> > >::state_case_type+0x82
000000d1`1e1f9e40 00007fff`4e050806 coreclr!__crt_stdio_output::output_processor<char,__crt_stdio_output::string_output_adapter<char>,__crt_stdio_output::format_validation_base<char,__crt_stdio_output::string_output_adapter<char> > >::process+0x34f
000000d1`1e1f9eb0 00007fff`4e051a41 coreclr!common_vsprintf<__crt_stdio_output::format_validation_base,char>+0x2a6
000000d1`1e1fa3e0 00007fff`4e06c4ba coreclr!common_vsprintf_s<char>+0x201
000000d1`1e1fa480 00007fff`4d77fb84 coreclr!__stdio_common_vsprintf_s+0x6a
(Inline Function) --------`-------- coreclr!_vsprintf_s_l+0x24
000000d1`1e1fa510 00007fff`4d762750 coreclr!sprintf_s+0x44
000000d1`1e1fa570 00007fff`4dd16679 coreclr!CONTRACT_ASSERT+0x210

[AaronRobinsonMSFT]

Looks like we have a problem with sprintf in CONTRACT_ASSERT macro:

Yes, that has been this way for a while. There are certain code paths that if they assert during a contract will cause a failure. The issue is a stale link in the contract stack frame. I am going to start on that soon.

[jkotas]

Dead code - delete? It does not seem to be called anywhere.

[jkotas]

I am not sure about this and other changed under debugger, and in eedbginterfaceimpl.cpp. Debugger often wants to do stuff in no-GC trigger mode. Have you been able to validate this with debugger tests on checked build of the runtime?

[AaronRobinsonMSFT]

@jkotas @jkoritzinsky I pushed up a bunch of random changes I made on the branch. I need to audit them still. Please ignore this PR for now.