Add SER308: flag the library's own Wait/WaitAll/TryWait helpers

[mgravell]

Follow-up to #3179, reworked after review feedback — the original approach used [Obsolete]; this one does not. The old commits are gone from the branch; the discussion above is the record of why.

Wait, WaitAll and TryWait are what SER307 discourages everywhere else, reached through the API the library itself offers for it: the calling thread is held for the round-trip while the reply needs a thread of its own to be processed. Shipping SER307 while providing a blessed way to do the same thing was only ever half a position.

Why an analyzer rule and not [Obsolete]

The feedback was right: [Obsolete] reports CS0618, which every obsoletion from every source shares. A consumer who wants to silence this has to silence all of them, including deprecations in their own code and in unrelated packages. That is far too blunt for an opinion we are choosing to have.

Options considered and rejected:

• ObsoleteAttribute.DiagnosticId — exists for precisely this, and would be the natural answer. It is net5+, and this library still targets netstandard2.0, so it would give a granular ID on some target frameworks and CS0618 on others.
• A hybrid (attribute where available, analyzer elsewhere) — the attribute would need #if on the interface and on ConnectionMultiplexer's own members, since marking only the interface does not warn through the class; and on modern targets both mechanisms would report at the same location.
• [Experimental] — granular, but these APIs are long-standing rather than preview, so it would be saying something untrue.

An ID of our own behaves identically on every target framework, sits with the rest of the SER3xx family, and has a docs page. Recorded in the descriptor as worth revisiting rather than settled: if netstandard2.0 and net4x are ever dropped, the attribute becomes strictly better — metadata, no analyzer needed, works in every tool — and SER308 can retire in its favour.

Two interfaces, not one

IRedisAsync declares these for database/server/subscriber calls; IConnectionMultiplexer declares its own, unrelated set. Both are gathered — testing only one would have left half the surface unguarded, which is exactly how the earlier attempt missed the multiplexer entirely.

Matching goes through FindImplementationForInterfaceMember as well as directly, because ConnectionMultiplexer is what Connect returns, making conn.Wait(task) the common shape with the class as the containing type.

Suppressions in this repo

Split into its own commit so the rule is reviewable on its own. Two kinds, distinguished in the comments rather than left to the reader:

• Forwarding — KeyPrefixed, RedisBase, the Retry*/MultiGroup* wrappers, the test fixture. They must implement the interfaces in full; forwarding a flagged member is not using it, and the implementation cannot be dropped while the interface declares it.
• One real one — CursorEnumerable's synchronous scan path, reached only from a caller who asked for the sync API, where TryWait is what applies the configured timeout.

Test call sites are suppressed rather than rewritten to await, deliberately: the Wait helpers apply the multiplexer's timeout and a bare await does not, so rewriting would quietly change what those tests exercise. Happy to convert them if preferred.

Verification

• 154 analyzer tests green, including negative cases: an unrelated Wait(Task) on someone else's type, Task.Wait(), and ManualResetEventSlim.Wait() are all left alone — the rule matches the interface members, not the name.
• Whole repo builds clean on all TFMs.
• Library test results unchanged vs main on the touched files.

[mgravell]

For @NickCraver and @philon-msft's consideration: I'm "done" with sync-over-async; thoughts on just saying no, but gently? See also https://seredis.dev/SyncOverAsync