Refactor Markdown length-limited summary implementation

[camelid]

This PR is a new approach to #79749.

This PR refactors the implementation of markdown_summary_with_limit(),
separating the logic of determining when the limit has been reached from
the actual rendering process.

The main advantage of the new approach is that it guarantees that all
HTML tags are closed, whereas the previous implementation could generate
tags that were never closed. It also ensures that no empty tags are
generated (e.g., <em></em>).

The new implementation consists of a general-purpose struct
HtmlWithLimit that manages the length-limiting logic and a function
markdown_summary_with_limit() that renders Markdown to HTML using the
struct.

r? @GuillaumeGomez

[GuillaumeGomez]

Overall, looks pretty nice. Please add tests to ensure it works as expected. Also, not sure to see the added value of ControlFlow in here...

Last point: we used to create the string with a capacity whereas you don't anymore.

[camelid]

Please add tests to ensure it works as expected.

Will do. 👍

Also, not sure to see the added value of ControlFlow in here...

The value is that it allows automatically stopping the iteration, and it communicates the intent better than Result (because it's not an error if it stops early) and better than bool (because bool doesn't work with try_for_each, and because it can be unclear whether values mean "stop" or "continue").

Or are you referring to the feature gate? Note that ControlFlow itself has recently been stabilized, and that feature gate is just for some convenience API like ControlFlow::BREAK = ControlFlow::Break(()).

Last point: we used to create the string with a capacity whereas you don't anymore.

See #88173 (comment).

[GuillaumeGomez]

The value is that it allows automatically stopping the iteration, and it communicates the intent better than Result (because it's not an error if it stops early) and better than bool (because bool doesn't work with try_for_each, and because it can be unclear whether values mean "stop" or "continue").

Or are you referring to the feature gate? Note that ControlFlow itself has recently been stabilized, and that feature gate is just for some convenience API like ControlFlow::BREAK = ControlFlow::Break(()).

No, I meant ControlFlow in general. But it's mostly a nit at this point.

[camelid]

I'll push up a fix for the test failures tomorrow.

[camelid]

Ok, CI passed, this should be ready for a new review!

[GuillaumeGomez]

Thanks!

@bors: r+

[bors]

📌 Commit 4478ecc has been approved by GuillaumeGomez