privacy: document unordered visibility limit

[SynapLink]

Follow-up to #155257.\n\nWhile investigating the FIXME in EffectiveVisibilities::update, the unordered case comes from restricted visibilities forming a tree: restrictions to sibling modules are incomparable. This documents the current behavior for the nominal visibility limit case, where the inherited effective visibility is preserved because there is no single Visibility that can represent the intersection.\n\nNo behavior change.

[rustbot]

r? @nikomatsakis

rustbot has assigned @nikomatsakis.
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

Why was this reviewer chosen?

The reviewer was selected based on:

• Owners of files modified in this PR: compiler
• compiler expanded to 73 candidates
• Random selection from 20 candidates

[petrochenkov]

r? @petrochenkov

[petrochenkov]

The FIXME is mostly for me, I'd like to find specific examples on which this code is triggered, whether they can be turned into errors, and what the update operation means for them if they are not errors.
A trivial LLM-generated explanation won't do.

[SynapLink]

@petrochenkov
The FIXME is mostly for me, I'd like to find specific examples on which this code is triggered, whether they can be turned into errors, and what the update operation means for them if they are not errors.
A trivial LLM-generated explanation won't do.

Im sorry. Ive used llm to summarize the whole thing, but i genuinely want to contribute to the repo, are there opened problems which needs to be solved or would you suggest some other thing, sorry for disturbing.