fix: split HTTP connect/read timeouts (Jellyfin build + qBit stats)

Every HTTP client passed an integer timeout to requests, applying the same
value to BOTH connect and read phases. A slow Jellyfin /Items page or qBit
/sync/maindata blew through the 10s read budget → ReadTimeoutError. Split
into a (connect=5s, read=60s default) tuple via shared http_timeout() helper.
The media index build worker uses a 180s read floor. Existing services with
low timeout_seconds benefit from bumping to 60+.

backend/tests/test_qbittorrent_client.py

@@ -129,6 +129,21 @@ class QbittorrentClientTests(unittest.TestCase):
         with self.assertRaises(requests.ConnectionError):
             client._login()
 
+    def test_qbittorrent_client_uses_tuple_timeout(self) -> None:
+        """The client passes a (connect, read) tuple to requests, not an int."""
+        # self.client was constructed with timeout=5 in setUp() and has a mocked session.
+        assert isinstance(self.client.timeout, tuple)
+        assert len(self.client.timeout) == 2
+        assert self.client.timeout[0] == 5.0  # connect timeout
+        assert self.client.timeout[1] == 5.0  # read timeout (what we passed)
+
+        # Verify it's actually passed to requests as-is.
+        self.session.post.return_value = self._login_response()
+        self.client._login()
+        call_kwargs = self.session.post.call_args.kwargs
+        assert call_kwargs["timeout"] == (5.0, 5.0)
+        assert not isinstance(call_kwargs["timeout"], int)
+
 
 if __name__ == "__main__":
     unittest.main()