Improve the `Ctrl+T` transcript viewer to allow sections to be 'collapsed', and then collapse overly verbose code reads/etc by default.

What feature would you like to see?

I would like the Ctrl+T transcript viewer to support collapsible sections within a transcript.

Specifically:

• Each "message" or "section" in the transcript (e.g., model responses, code reads, context blocks) should be individually collapsible/expandable.
• By default, overly verbose sections—such as long code reads—should start in a collapsed state.
• Users could expand them on demand if they want to verify the exact details, while keeping the transcript view concise for normal navigation.

This would preserve transparency (the data is still there and accessible) while improving usability (avoiding the overwhelming wall of text that long code reads can generate).

Why is this needed?

Currently, transcripts that include large code read operations become difficult to navigate because those sections dominate the view. While these reads can be useful for transparency (understanding what the model had in context), in most day-to-day debugging or review they’re too noisy and make the transcript hard to skim.

Having collapsible sections solves both sides:

• Usability: reduces transcript clutter and keeps the focus on higher-level interactions.
• Transparency: users can still expand the code if they need to verify exactly what was read.

Alternative considered

One idea raised was to simply remove code reads entirely, replacing them with a one-line summary like:

Read from App.tsx (lines 10–123)

While this would simplify things, it may lose useful detail in situations where seeing the actual read context matters. Collapsing provides a more flexible middle ground.

Are you interested in implementing this feature?

Unfortunately I don't have the capacity or familiarity with rust to take this on right now.

Additional information

This feature request came out of feedback that transcripts are currently cluttered by code reads. Collapsing by default, while still allowing expansion, balances visibility with usability:

> I think I would be fine with this change, but I think they should remove the code reads from the transcript because I don't see how they are useful. Maybe just keep a one-line like _Read from App.tsx from line 10 to 123_, or whatevr @sldx I think they are useful sometimes in that they provide transparency for the context the model was seeing; but I agree that in a lot of standard usecases they end up being overly verbose. I think it would be nice if the Ctrl+T transcript viewer allowed 'messages' / 'sections' of the transcript to be 'collapsed'; that way longer code view sections could be collapsed by default, but we could select them and expand it if we really wanted to see what was in them. Though I suspect creating a more focussed issue about that for visibility would be better than it being lost in these comments. _Originally posted by @0xdevalias in #2375_