Generate video from text prompt

[hipsterusername]

Summary

This PR introduces a new "Video" feature, including a dedicated Video Launchpad, gallery, and management capabilities. The changes encompass:

• Backend API: New FastAPI routers for video operations (CRUD, starring, board association, listing).
• Backend Model Management: Integration of video models (Veo3, Runway) into the model manager.
• Frontend UI:
  • A new "Video" tab in the main application.
  • A "Video Launchpad" panel with the requested text updates.
  • Video gallery components, a video viewer, and video-specific context menus and actions.
  • Integration of video items into existing gallery, board, and drag-and-drop logic.
  • Addition of necessary i18n strings and third-party video playback libraries.
• Workflow Integration: VideoField and VideoOutput for nodes in the workflow editor.
• Canvas Integration: Canvas aspect ratio and dimensions now update when a video model is selected.

Related Issues / Discussions

This work stems from a Slack discussion regarding the implementation of the "Video Launchpad" UI and associated copy updates.

QA Instructions

This PR focuses on implementing the core video feature and frontend UI. The backend API endpoints are currently stubs (raise HTTPException(status_code=501, detail="Not implemented")), so full end-to-end functionality for video generation and persistence is not yet available. Per user instruction, this PR has not been built or tested by the AI.

To QA the frontend changes:

1. Verify the new "Video" tab is present in the main application.
2. Navigate to the "Video" tab and observe the "Video Launchpad" panel.
3. Confirm the updated UI text:
   • "Generate videos from text prompts."
   • "Select Your Model" (dropdown placeholder)
   • "Want to learn what prompts work best for each model? Check out our Model Guide."
   • "Generate from Text"
   • "Enter a prompt and Invoke."
   • "Add a Starting Frame"
   • "Add an image to control the first frame of your video."
4. Check the new "Video" gallery view (it will be empty without backend data).
5. Verify that the "Add a Starting Frame" button is present and functions as an image drop target.
6. Observe that the canvas aspect ratio and dimensions update to 16:9 (1280x720) when a video model is selected in the Video Launchpad.

Merge Plan

This PR introduces a significant new feature. The backend API endpoints are currently stubs and will require further implementation. Frontend changes are extensive and touch many core components. Coordination with backend development for full video functionality will be required.

Checklist

• The PR has a short but descriptive title, suitable for a changelog
• Tests added / updated (if applicable)
• Documentation added / updated (if applicable)
• Updated What's New copy (if doing a release after this PR)

---

Slack Thread

[cursor]

Cursor Agent can help with this pull request. Just @cursor in comments and I'll start working on changes in this branch.
_{Learn more about Cursor Agents}