Enable per-conversation loading states to allow having parallel conversations

[allozaur]

Close #16133

Introduces granular loading state management for individual conversations, allowing concurrent message processing which prevents UI lockup and improves user experience when multiple conversations are active.

[ggerganov]

Would it be possible to push the new index?

[allozaur]

@ggerganov here you go (17e80c5) 😉

[ggerganov]

Noticed 2 issues:

• The "regenerate" button stops working after I switch to another conv
• The displayed stats do not correspond to the selected conv while another conv is being generated

[ServeurpersoCom]

LLama-Svelte-Parallel-4-AVC500Kbps.mp4

[allozaur]

Noticed 2 issues:

* The "regenerate" button stops working after I switch to another conv

* The displayed stats do not correspond to the selected conv while another conv is being generated

Yep. Will fix these

[ggerganov]

I have a small feature request about this feature if possible:

The list of conversation on the left should have some indicator about which conversations are currently being generated.

[allozaur]

I have a small feature request about this feature if possible:

The list of conversation on the left should have some indicator about which conversations are currently being generated.

Ah, of course, that's a very reasonable request :)