Connector listing fails silently with no error logging or UI indication #1963

Issue submitter TODO list

• I've looked up my issue in FAQ
• I've searched for an already existing issues here
• I've tried running main-labeled docker image and the issue still persists there
• I'm running a supported version of the application which is listed here

Describe the bug (actual behavior)

When the connector listing fails, kafbat UI renders an empty list and logs nothing.

api/src/main/java/io/kafbat/ui/service/KafkaConnectService.java:195-197:

// returns empty flux if there was an error communicating with Connect
public Mono<Map<String, ExpandedConnector>> getConnectorsWithErrorsSuppress(
    KafkaCluster cluster, String connectName) {
  return getConnectors(cluster, connectName).onErrorComplete();
}

onErrorComplete() discards the error before anything can log or display it. The class is annotated
@Slf4j, but the logger is never used on this path.

Expected behavior

At minimum, I believe the error should be logged.

Your installation details

1. App version -> dfa5a7e
2. Config format

kafka:
  clusters:
    - name: <cluster>
      bootstrapServers: <redacted>
      kafkaConnect:
        - name: <connect-cluster>
          address: http://<connect-host>

Steps to reproduce

1. Register a Kafka Connect cluster in kafbat UI (kafka.clusters[].kafkaConnect[]).
2. Register a connector whose connector.class is not present in the worker image — e.g.
   io.confluent.connect.s3.S3SinkConnector on a worker without the S3 plugin. In the wild this
   happens when a worker image upgrade drops a plugin while the connector config remains in the
   config storage topic.
3. Keep at least one healthy connector in the same Connect cluster.
4. Confirm the asymmetry against the Connect REST API directly:

$ curl -s <connect>/connectors
["bigquery-connector","s3-sink-connector"]     # 200 - both are registered and listed, because
                                               # names alone need no class loading.
                                               # note: the S3 plugin is NOT in the worker image,
                                               # yet s3-sink-connector still shows up here

$ curl -s -o /dev/null -w "%{http_code}\n" '<connect>/connectors?expand=info&expand=status'
500                                            # the exact same two connectors, but each entry now
                                               # carries `type`, so every class must resolve -
                                               # s3-sink-connector's cannot, and that one failure
                                               # takes down the entire response

5. Open the Connect tab for that cluster in kafbat UI.

Result: the list is empty, as if the cluster had no connectors at all. The healthy
bigquery-connector is not shown either. Nothing appears in the application log.

Screenshots

No response

Logs

No response

Additional context

the underlying fault is on the Kafka Connect side (a stale connector registration whose plugin is no longer in the worker image), and the all-or-nothing behaviour of the expanded listing endpoint is Connect's own design. If you consider this something that belongs in Connect rather than in kafbat UI, please feel free to close this issue.