Update entire CImageTreeModel table on compression finish.

[AlexeyPeov]

The images table now does not update all of its contents when compression finishes. It redraws only when you put a cursor over it (1st video). I propose a quick fix, to emit dataChanged on compression finish. 2nd video demonstrates that the issue is gone.

Before fix: https://github.com/user-attachments/assets/0e1ca118-06cd-4cea-93e3-79ac1c2d38d8

After fix: https://github.com/user-attachments/assets/d2844394-e5a5-4766-ab3a-fb94f2201ddd

[Lymphatus]

Good catch, I had to watch a few times the two videos to get what you meant 😄
My only concern is that it can affect performance on very large lists. Did you notice anything on that regard?

PS: pipelines failing for unrelated reasons.

[AlexeyPeov]

About the pipelines - yeah, I had to build the rust lib manually + remove language translation files from "/resources/languages.qrc" to make the project compile (compiler complained about "circular dependencies"). Kinda fixed that, but decided not to include it in this pr.

Now on the topic of performance - there are 3 ways of thinking about it:

1st (my solution): I figured that updating the table manually on compression finish is exactly what happens when you put a cursor over it (the "CImageTreeModel::dataChanged" fn just triggers HtmlDelegate::paint, the performance impact is insignificant, correct me if I'm wrong). I put a counter on HtmlDelegate::paint, and it measured ~700 repaints on compression stage both before and after the change, I used 25 images.

2nd: do not trigger "dataChanged" from "QFutureWatcherBase::progressValueChanged", and only trigger it in "QFutureWatcherBase::finished", this would result in the same amount of calls of "dataChanged" as was before, but it will make compression visually less responsive (bad from user experience standpoint imo).

3rd (this I hadn't figured out): Somehow have the correct status in CImage::getFormattedStatus when QFutureWatcherBase::progressValueChanged for an image gets called, the would be ideal, but I'm not that familiar with this codebase yet, and if you know the solution please let me know.

TLDR: I decided to take the path of least resistance =)

[Lymphatus]

Your solution sounds right to me. Thanks!