Add Bubble Tea TUI migration investigation document

[Copilot]

Investigation of migrating the TUI from Python/Textual to Go/Bubble Tea, prompted by PR #2's hybrid implementation.

Findings

• Current state: 526 LOC Textual TUI, 24% test coverage, known bugs (duplicate on_button_pressed, missing main())
• Performance gap: Textual startup 2-5s (ML imports), Bubble Tea <100ms; memory 500MB+ vs ~50MB

Recommendation

Proceed with hybrid approach from PR #2:

• Go/Bubble Tea frontend for rendering and input
• Python CLI backend (src/cli.py) for Reddit API and ML features via subprocess + JSON

Deliverable

docs/BUBBLE_TEA_MIGRATION_INVESTIGATION.md covers:

• Current architecture analysis
• Framework comparison (Textual vs Bubble Tea)
• Three migration options evaluated
• PR Add Bubble Tea TUI and CLI bridge #2 implementation review
• Risk analysis and mitigation
• Phased implementation roadmap (2-4 days estimated)

Key Tradeoffs

Approach	Startup	Memory	Complexity
Keep Textual	2-5s	500MB+	Low
Hybrid (PR #2)	<100ms	50MB	Medium
Full Go rewrite	<100ms	50MB	High (loses Python ML)

Original prompt

INVESTIGATION of a bubble tea migration for the TUI

---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.