Move usages of COMPlus_ var to DOTNET_ fallback

[am11]

Following up on #77025 for the remaining COMPlus_ usages in src/coreclr/{tools,nativeaot}.

[dotnet-issue-labeler]

I couldn't figure out the best area label to add to this PR. If you have write-permissions please help me learn by adding exactly one area label.

[BruceForstall]

Looks good except for one thing.

[am11]

win-x64 test failure is unrelated gc assertion; #76801.

@AaronRobinsonMSFT, what do you think about issuing an obsoletion notice as "warn once" (using a static variable) in this block:

runtime/src/coreclr/utilcode/clrconfig.cpp

Lines 661 to 664 in 4cbe6f9

	&& SString::_wcsnicmp(wszName, COMPLUS_PREFIX, LEN_OF_COMPLUS_PREFIX) == 0)
	{
	wszName += LEN_OF_COMPLUS_PREFIX;
	s_EnvNames.Add(wszName, (DWORD) (wszCurr - wszName));

Warning: COMPlus_ prefixed environment variables will be removed in the future, use DOTNET_ prefix instead.

By net10.0, we can remove the COMPlus_ fallbacks and issue an error if COMPlus_ variables are set in the environment (or change the code to nop and issue a deprecation warning: `COMPlus_` variables has no affect. Use `DOTNET_` instead.).

[AaronRobinsonMSFT]

@AaronRobinsonMSFT, what do you think about issuing an obsoletion notice as "warn once" (using a static variable) in this block:

All for it but understanding how that works in practice is hard. It was in my original proposal for this work actually. Start reading at #47283 (comment). Unfortunately, I deleted the gist I wrote my proposal in, but the idea is captured in those comments.

[BruceForstall]

@BruceForstall, could you please take another look?

I'll try to look soon.

[BruceForstall]

Thanks for all the work.