Save chat prompt to thread before the AI responds

[shatfield4]

Pull Request Type

• ✨ feat (New feature)
• 🐛 fix (Bug fix)
• ♻️ refactor (Code refactoring without changing behavior)
• 💄 style (UI style changes)
• 🔨 chore (Build, CI, maintenance)
• 📝 docs (Documentation updates)

Relevant Issues

resolves #6718

Description

• save the user's prompt to the thread before the LLM is called so it is kept if the response fails or the user leaves mid-generation
• applies to both regular workspace chats and agent chats
• unanswered prompts stay visible in the thread and can be re-sent with edit + submit
• unanswered prompts are left out of the history sent to the LLM since some providers reject a blank assistant message

Visuals (if applicable)

Additional Information

• LLM unreachable → error shown, prompt still in thread after reload
• invalid model (API error) → prompt still in thread after reload (lost on master)
• reload before the first token → prompt still in thread after reload
• reload mid-stream → prompt + partial response saved
• follow-up chat after a failed prompt works and the chat row is updated in place (no duplicate)
• edit + submit on a failed prompt replaces it and gets a response

Developer Validations

• I ran yarn lint from the root of the repo & committed changes
• Relevant documentation has been updated (if applicable)
• I have tested my code functionality
• Docker build succeeds locally

[timothycarambat]

If you send a prompt successfully in an agent session and then force fail it on the second - does the second submitted prompt save on reload?