[Dot Shorthands] Dartdoc Support

[kallentu]

This issue tracks the work needed for supporting dot shorthands in Dartdoc.

Estimated Work

• Confirm that [.id] or [.id()] should not work (since there's no context available)

I don't see how else this could be used, not sure. Will add as we think of tasks.

cc. @srawlins

[srawlins]

SG. Usually these expression-level features are straightforward in dartdoc.

[sigurdm]

Is the implementation far enough ahead at main, that we can test this already?

[kallentu]

Is the implementation far enough ahead at main, that we can test this already?

So far as testing that [.id] and [.id()] don't work, I believe we are ready. There's no context type so it shouldn't resolve.

The relevant parts of the analyzer AST (DotShorthandPropertyAccess and DotShorthandMethodInvocation) is ready in the latest versions. The Dartdoc comment references shouldn't resolve either way.

[srawlins]

Two other areas that might be affected:

• annotations, like @Foo(month: .january) - we don't link .january (though I wish we did; there is an open issue), but we should make sure we don't crash.
• parameter default values, like void m({Month moth = .january}) {} - we again don't link, I think, but we should make sure we don't crash.

[kallentu]

As of analyzer 8.0.0, everything should be ready for testing (.id, .id() as a method or a constructor).
@sigurdm, circling back, do we have anyone that can tackle the testing for this?

[sigurdm]

@szakarias might have the time next week. I guess we could make some automated test cases for this, so we don't happen to regress.

[kallentu]

[@szakarias](https://github.com/szakarias) might have the time next week. I guess we could make some automated test cases for this, so we don't happen to regress.

Great, thank you both.

[szakarias]

Added tests to make sure we don't crash. In #4093 , #4092, and #4089 .