fix(keyword_docs.rs): use of as keyword to disambiguate paths

[iampi31415]

• The file keyword_docs.rs was missing documenting the use of the as keyword to disambiguate paths in trait implementations.

This is also described in the issue linked below.

There isn't a tracking issue, but there is an issue here.

Some details:

• I could include a link to the RFC section https://github.com/rust-lang/rfcs/blob/master/text/0132-ufcs.md#paths-that-begin-with-an-assoc_segment, if you think that it's appropriate.
• The style was inspired by the for docs

[rustbot]

r? @ChrisDenton

rustbot has assigned @ChrisDenton.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

[fmease]

Coincidentally someone already opened a PR a week ago: #142670.

Not sure if we should close this PR in favor of the other one or if we should wait for the older one to get merged first and then recontextualize this PR to act as a follup up / fixup.

Anyway @rustbot blocked

[iampi31415]

@fmease I hadn't noticed, that's bad luck. From what I can tell this fix seems more complete and quite different content. I don't know what's the best path forward.

[iampi31415]

Current built page looks like this:

[tshepang]

can you squash these commits, given that most of them are related fixups

[iampi31415]

can you squash these commits, given that most of them are related fixups

@tshepang yes. Is that okay? (should be a single commit now.)

[RalfJung]

Please give this PR a description that summarizes what happens here. "I fixed something in that file" is not super helpful :) Please mention at least the 'as' keyword.

[iampi31415]

@RalfJung thanks, sorry about that. Is now fine or still not ideal?

I changed both description and title.

[fpdotmonkey]

So I wrote the other PR. My opinion of resolving this conflict would be to treat my contribution as the textual description of how this syntax works and this as a detailed example.

My doc goes into some depth on when you actually want to use this syntax and provides a (slightly bad) real-world example. This PR offers a more detailed, but contrived example. These combined I think can harmonize quite well.

[fpdotmonkey]

Also, would it make more sense to illustrate this concept with a macro definition than a simple method call? Macros are probably a more practical and common use case than actually disambiguating an associated item. Granted, macros bring in a bunch of conceptual overhead that might not be desirable.