The current users.json dataset is ~6MB for ~6k users (~1KB/user). Scaling to 100k users would result in an unmanageable ~100MB payload.

We need to drastically optimize the data structure to reduce file size by ~70% while maintaining functionality.

Optimization Strategy:

1. Short Keys (Mapping): Map verbose keys to short codes (e.g., login -> l, total_contributions -> tc).
2. Avatar IDs: Store only the integer ID for avatars (User & Orgs), not full URLs.
3. Years Array: Convert the years object map to a sequential array starting from first_year.
4. Org Simplification: Limit orgs to top 5 and store as a compact tuple [login, id].
5. Smart Formatting: Use a "One-Record-Per-Line" format (valid JSON array, but each object is minified on a single line) to balance compression with git-diffability.

Tasks:

• Update DevRank.model.Contributor with field mappings and reconstruction logic.
• Update DevRank.services.Updater to persist data in this new format.
• Migrate existing users.json to the new schema.