Three-tier constraint model, mode-aware eval, boundary examples, playtest tooling

Eval harness:
- Mode-aware scoring: sudo=strict (exact match), pray/god=soft (category match,
  in-character, appropriate intensity)
- New metrics: cmd_category_match, appropriate_intensity, scoring_mode breakdown
- Eval defaults to steel141 (192.168.0.141) — prod GPU reserved for serving

Dataset (213 examples):
- Added 31 boundary/adversarial examples (safety edges, abstention, near-boundary)
- Updated pray example reasoning: character-driven logic, not prescriptive outputs
- Tagged pray examples with scoring_mode=soft

Playtest tooling:
- whitelist.sh: add/remove/list across all 3 servers
- FRIENDS_INVITE.md + Discord version: playtester recruitment docs
- Server addresses and implementation details for both training servers

PLAN.md:
- Three-tier constraint model documented (sudo/pray/god_system)
- Success criteria split by scoring mode
- All session decisions logged

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

eval/bakeoff.py

@@ -317,7 +317,7 @@ def run_bakeoff(models: list, ollama_url: str, no_think: bool = False):
 
 def main():
     parser = argparse.ArgumentParser(description="Model Bake-Off")
-    parser.add_argument("--ollama-url", default="http://192.168.0.179:11434")
+    parser.add_argument("--ollama-url", default="http://192.168.0.141:11434")
     parser.add_argument("--models", nargs="+",
                         default=["qwen3-coder:30b", "gemma3n:e4b"])
     parser.add_argument("--no-think", action="store_true",