Add rulebook RAG pipeline and LLM-driven game setup wizard

RAG: switch Postgres to pgvector, chunk and embed the three D&D
rulebooks locally via sentence-transformers, and retrieve relevant
excerpts per DM turn (query = latest player message) to ground the
system prompt. Retrieval runs off the event loop and is capped by a
relevance threshold and a max character budget so it can't blow up
context size or cost.

Game setup wizard: creating a game now opens a short chat where the
DM asks about genre, length, and the player's experience level, then
proposes a name and description via a tool call. The player can edit
both before creating the game. Stateless endpoint — the frontend
carries the conversation, no DB needed since the game doesn't exist
yet.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

backend/app/llm/game_setup.py

@@ -0,0 +1,55 @@
+import json
+
+from app.config import settings
+from app.llm.client import get_game_setup_prompt, get_llm_client
+from app.llm.tools import game_setup as game_setup_tool
+
+MAX_ROUNDS = 3
+MAX_TOKENS = 1024
+
+TOOLS = [{"type": "function", "function": game_setup_tool.TOOL_SCHEMA}]
+
+
+async def run_game_setup_turn(client_messages: list[dict]) -> dict:
+    """Stateless setup-wizard turn: the caller owns conversation history (no DB, no game yet).
+
+    Returns {"messages": <history to resend next turn>, "assistant_text": str | None,
+    "proposal": {"name": str, "description": str} | None}.
+    """
+    client = get_llm_client()
+    messages = [{"role": "system", "content": get_game_setup_prompt()}] + client_messages
+
+    assistant_text: str | None = None
+    proposal: dict | None = None
+
+    for _ in range(MAX_ROUNDS):
+        response = await client.chat.completions.create(
+            model=settings.dm_model,
+            max_tokens=MAX_TOKENS,
+            messages=messages,
+            tools=TOOLS,
+        )
+        message = response.choices[0].message
+        if message.content:
+            assistant_text = message.content
+
+        if not message.tool_calls:
+            messages.append({"role": "assistant", "content": message.content or ""})
+            break
+
+        messages.append(message.model_dump(exclude_unset=True))
+        for tool_call in message.tool_calls:
+            args = json.loads(tool_call.function.arguments)
+            if tool_call.function.name == "propose_game_setup":
+                proposal = {"name": args.get("name", ""), "description": args.get("description", "")}
+            messages.append(
+                {
+                    "role": "tool",
+                    "tool_call_id": tool_call.id,
+                    "content": json.dumps({"status": "ok"}),
+                }
+            )
+        if proposal:
+            break
+
+    return {"messages": messages[1:], "assistant_text": assistant_text, "proposal": proposal}