fix: JSON repair, WS thread pool, switch to DeepSeek V4 Flash, funnel Lv0

LLM fixes:
- _try_repair_json() — ripara JSON malformato (apici, virgole trailing)
- _strip_json_from_text() — rimuove residui JSON dalla narration
- parse_tool_call() — rimuove troncamento 200 caratteri dal fallback

DM fixes:
- Fallback euristico quando params vuoti (dm.py)
- Estrazione narrazione da dm_override/narrate params.params.text

Prompt (prompts.py):
- Vietati termini tecnici/scientifici/anacronistici (HPLC, laser, ...)
- Aggiunto vocabolario fantasy obbligatorio

Server (server.py):
- dm_cycle in ThreadPoolExecutor per non bloccare WS ping
- WebSocketDisconnect gestiti
- Rimosso import inesistente db_schema

Frontend (index.html):
- Nascosti tool name dalla chat
- Personaggio -> PG, connessioni aree con nomi

Infrastruttura:
- Switch da openrouter/free a DeepSeek V4 Flash via OpenCode Go
- pyproject.toml, Dockerfile, docker-compose.yml, start.sh
- .env auto-load (start.sh)
- Funnel Lv0 (funnel.py) — creazione PG livello 0

Docs: architecture, database, setup, game mechanics, codebase ref

src/engine/config.py

@@ -15,10 +15,10 @@ class ProviderConfig:
     openrouter_model: str = "openrouter/free"
     openrouter_key_env: str = "DCC_OPENROUTER_KEY"
     openrouter_free_fallback: bool = True
-    local_base_url: str = "http://192.168.0.240/v1"
+    local_base_url: str = "https://opencode.ai/zen/go/v1"
     local_api_key: str = ""
-    local_model: str = "big-pickle"
-    timeout: int = 60
+    local_model: str = "deepseek-v4-flash"
+    timeout: int = 120
 
 
 def load_from_env() -> ProviderConfig: