Add aspect ratio and image size options to image generation; implement chat-based image generation handling

Co-authored-by: Copilot <copilot@github.com>

backend/app/services/openrouter.py

@@ -139,3 +139,34 @@ async def poll_video_status(polling_url: str) -> dict[str, Any]:
         response = await client.send(resp)
         response.raise_for_status()
         return response.json()
+
+
+async def generate_image_chat(
+    model: str,
+    prompt: str,
+    modalities: list[str] | None = None,
+    image_config: dict[str, Any] | None = None,
+) -> dict[str, Any]:
+    """Request image generation via Chat Completions with modalities.
+
+    Used by models like FLUX.2 Klein 4B and GPT-5 Image Mini that output
+    images through the chat completions endpoint rather than /images/generations.
+    """
+    base_url = os.getenv("OPENROUTER_BASE_URL", OPENROUTER_BASE_URL)
+    if modalities is None:
+        # Image-only models (FLUX) vs multimodal (GPT-5 Image Mini)
+        modalities = ["image"]
+    payload: dict[str, Any] = {
+        "model": model,
+        "messages": [{"role": "user", "content": prompt}],
+        "modalities": modalities,
+    }
+    if image_config:
+        payload["image_config"] = image_config
+    async with httpx.AsyncClient(timeout=120) as client:
+        resp = client.build_request(
+            "POST", f"{base_url}/chat/completions", headers=_headers(), json=payload
+        )
+        response = await client.send(resp)
+        response.raise_for_status()
+        return response.json()