Add UNO game, hangman threading, DM room reuse, and gameplay fixes

- Add full UNO game plugin (1v1 vs bot via DM, community pot, bot personality)
- Move hangman into threads with direct reply guessing (no !hangman prefix needed)
- Fix DM room duplication by checking m.direct account data with in-memory cache
- Auto-draw when player has no playable cards instead of requiring manual draw
- Remove forced UNO call phase — players are penalized naturally if they forget
- Fix mutex held during network I/O in hangman start
- Fix infinite loop when both sides have no playable cards and deck is empty
- Fix Wild Draw Four penalty being skipped when UNO prompt triggered
- Fix hangman displayPhrase word boundary clarity (triple-wide gaps for spaces)
- Add UNO env vars, db tables, and README updates

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

README.md

@@ -285,7 +285,7 @@ Rep is earned when someone thanks you. The bot detects this automatically.
 |---------|-------------|
 | `!flip` | Coin flip |
 | `!games` | List available games |
-| `!hangman start` | Start a Hangman game |
+| `!hangman start [easy\|medium\|hard\|extreme]` | Start a Hangman game (optional difficulty) |
 | `!hangman [letter]` | Guess a letter |
 | `!hangman [phrase]` | Attempt full solution |
 | `!hangman submit [phrase]` | Submit a phrase to the pool |
@@ -450,6 +450,7 @@ All of these run in the background without any commands:
 - **Keyword alerts** - DMs you when someone says your watched keywords
 - **Presence** - auto-clears away/afk when you send a message
 - **Room milestones** - announces at 1k, 5k, 10k, 25k, 50k, 100k, 250k, 500k, 1M messages
+- **Euro earning** - earn €0.50–€10 per message based on word count (30s cooldown). Starting balance seeded from corpus history (capped at €2,500)
 - **URL previews** - OG tag extraction (feature-flagged, off by default)
 - **Reactions** - logs all reactions for `!emojiboard`
 - **Space groups** - rooms with sufficient member overlap are automatically grouped. Leaderboards, trivia scores, and other per-room features aggregate across the group. Recomputed hourly; persisted to SQLite. Uses strict clique-based grouping (every room must meet the threshold with every other room in the group).