Add scoreboard roast feature via image analysis

When @mentioned with an image attachment, the bot now roasts players
based on scoreboard screenshots using the vision model. Text-only
mentions continue to work as before.

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

utils/llm_client.py

@@ -1,4 +1,5 @@
 import asyncio
+import base64
 import json
 import logging
 from pathlib import Path
@@ -238,6 +239,55 @@ class LLMClient:
                 logger.error("LLM chat error: %s", e)
                 return None
 
+    async def analyze_image(
+        self,
+        image_bytes: bytes,
+        system_prompt: str,
+        user_text: str = "",
+        on_first_token=None,
+    ) -> str | None:
+        """Send an image to the vision model with a system prompt.
+
+        Returns the generated text response, or None on failure.
+        """
+        b64 = base64.b64encode(image_bytes).decode()
+        data_url = f"data:image/png;base64,{b64}"
+
+        user_content: list[dict] = [
+            {"type": "image_url", "image_url": {"url": data_url}},
+        ]
+        if user_text:
+            user_content.append({"type": "text", "text": user_text})
+
+        async with self._semaphore:
+            try:
+                stream = await self._client.chat.completions.create(
+                    model=self.model,
+                    messages=[
+                        {"role": "system", "content": system_prompt},
+                        {"role": "user", "content": user_content},
+                    ],
+                    temperature=0.8,
+                    max_tokens=500,
+                    stream=True,
+                )
+
+                chunks: list[str] = []
+                notified = False
+                async for chunk in stream:
+                    delta = chunk.choices[0].delta if chunk.choices else None
+                    if delta and delta.content:
+                        if not notified and on_first_token:
+                            await on_first_token()
+                            notified = True
+                        chunks.append(delta.content)
+
+                content = "".join(chunks).strip()
+                return content if content else None
+            except Exception as e:
+                logger.error("LLM image analysis error: %s", e)
+                return None
+
     async def raw_analyze(self, message: str, context: str = "", user_notes: str = "") -> tuple[str, dict | None]:
         """Return the raw LLM response string AND parsed result for /bcs-test (single LLM call)."""
         user_content = f"=== CONTEXT (other users' recent messages, for background only) ===\n{context}\n\n"