[Console] Fix four DTO/entity defects that break Spark and Flink SQL app management

[88fantasy]

What is the purpose of the change

Four small DTO/entity defects, each of which makes a normal UI operation fail outright on a clean install. They are unrelated to each other in cause but identical in shape — a field that does not line up between the request DTO, the entity, and the column — so they are grouped in one PR.

Closes #4491
Closes #4492
Closes #4493
Closes #4494

Brief change log

1. FlinkAppCreateRequest declares sqlId ([Bug] Saving an already-started Flink SQL application from the UI always fails (FlinkAppCreateRequest has no sqlId) #4491). The frontend sends it and updateFlinkSqlJob() requires it, but the DTO had no such field, so BeanUtils.copyProperties skipped it and saving any already-started Flink SQL application failed with Flink sql is null, update flink sql job failed. SparkAppCreateRequest and FlinkAppResponse already declare it — this was a one-sided omission.

2. SparkAppStateEnum.of(Integer) returns OTHER for null ([Bug] Spark application list page returns 500 (SparkAppStateEnum.of unboxes a null state) #4492). It compared with appState.value == state, which unboxes; SparkAppListQueryRequest's scalar state is null for an unfiltered list, so /spark/app/list returned 500 for everyone. FlinkAppStateEnum.getState(Integer) has the same unboxing shape but nothing reaches it with a null today, so it is deliberately left alone here.

3. SparkEnv.doSetSparkConf() stores an empty conf when the file is absent ([Bug] A stock Spark distribution cannot be registered (spark-defaults.conf is absent, spark_conf is NOT NULL) #4493). A stock Spark distribution ships only spark-defaults.conf.template, so sparkConf stayed null and the insert failed on t_spark_env.spark_conf (NOT NULL, no default) — no Spark home could be registered at all.

4. SparkApplication.k8sImagePullPolicy is Integer ([Bug] Creating a Spark application from the UI always returns 500 (k8sImagePullPolicy primitive int) #4494), matching its nullable column and the Integer in SparkAppCreateRequest. As a primitive it unboxed the null the Spark frontend never sends, so /spark/app/create returned 500 for every UI-created application.

Verifying this change

• (1) and (2) were verified against a running console: the same update request goes from always-failing to succeeding, and /spark/app/list from 500 to 200.
• (3) was verified by the inverse — creating an empty conf/spark-defaults.conf in an otherwise stock Spark home makes registration succeed with the file's existence as the only changed variable.
• mvn test on streampark-console-service passes (93 tests, 0 failures).

Does this pull request potentially affect one of the following parts

• Dependencies (does it add or upgrade a dependency): no
• The public API: yes, minor — FlinkAppCreateRequest gains an optional sqlId, and SparkApplication.getK8sImagePullPolicy() now returns Integer rather than int.
• The runtime per-record code paths (performance sensitive): no
• Anything that affects deployment: no

Documentation

• Does this pull request introduce a new feature? no

[sonarqubecloud]

Quality Gate passed

Issues
1 New issue
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[wolfboys]

I noticed your recent code contributions — they look good. I wanted to ask which AI model you used, including the specific model/version, and how you prompted the model to complete this contribution.

We’re currently planning to introduce some guidelines and best practices for AI-assisted coding in our open-source project, so your experience would be very helpful.