Cache room and message data on disk for faster startup

[sgrimee]

Summary

Implement persistent disk caching for room and message data to significantly reduce startup time by avoiding the need to re-fetch all data from the server on each launch.

Requirements

• Cache room metadata and message history to disk
• Load cached data on startup for immediate UI population
• Update cache when loading fresh data from server
• Maintain message ordering and structure integrity
• Handle cache invalidation when server data changes

Research Needed

• Investigate Rust best practices for data serialization/storage (serde, bincode, JSON, etc.)
• Evaluate performance trade-offs of different storage formats
• Consider cache size management and cleanup strategies

Implementation Considerations

• Cache may contain more message history than typical API calls (due to pagination)
• Must preserve message ordering and threading structure
• Leverage existing functions that ensure messages are inserted correctly
• Integrate with existing cache system in src/app/cache/
• Handle concurrent access to cache files
• Consider cache versioning for schema changes

Critical Requirements

• Message structure correctness must be maintained
• Cache updates should be atomic to prevent corruption
• Graceful fallback when cache is corrupted or missing
• Efficient merge of cached and fresh data from server

Acceptance Criteria

• Room and message data persists between sessions
• Startup time significantly reduced with cached data
• Cache updates correctly when new data arrives from server
• Message ordering and structure integrity maintained
• Cache handles pagination and history correctly
• Performance acceptable for large message histories

🤖 Generated with Claude Code