feat: add Chinese UI, Responses models, and mirrored Windows setup

Localize the web console and persist custom model and capability settings.
Use OpenAI Responses exclusively and move runtime limits into environment configuration.
Bootstrap uv and the virtual environment on Windows, support optional China mirrors,
and package source distributions without environments, build output, or credentials.

Update regression tests and repository quality checks for the new behavior.

playground/backend_manager/app/auth/jwt_handler.py

@@ -12,7 +12,7 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-from datetime import datetime, timedelta, timezone
+from datetime import datetime, timedelta, timezone, UTC
 from fastapi import Depends, HTTPException, Security, status
 from fastapi.security import HTTPAuthorizationCredentials, HTTPBearer
 from jose import JWTError, jwt
@@ -23,13 +23,11 @@ security = HTTPBearer(auto_error=False)
 
 def create_access_token(user_id: str, extra_claims: dict | None = None) -> str:
     """Generate a signed JWT access token."""
-    expire = datetime.now(timezone.utc) + timedelta(
-        minutes=settings.JWT_ACCESS_TOKEN_EXPIRE_MINUTES
-    )
+    expire = datetime.now(UTC) + timedelta(minutes=settings.JWT_ACCESS_TOKEN_EXPIRE_MINUTES)
     to_encode = {
         "sub": user_id,
         "exp": expire,
-        "iat": datetime.now(timezone.utc),
+        "iat": datetime.now(UTC),
         "iss": "artemis-backend-manager",
     }
     if extra_claims: