Add kafka,json and hadoop-mapreduce-client-core jar in flink engine

[saLeox]

What is the purpose of the change

1. Fix a ClassNotfoundException issue because of lacking hadoop-mapreduce-client-core dependency in Flink engine pom;
2. Since Kafka and Json are the more usual used in Flink streaming use case, it's necessary to add on in pom, in case user use them in Flink SQL.

Brief change log

• Add kafka, json and hadoop-mapreduce-client-core jar in flink engine pom.

Verifying this change

This change is already covered by existing tests.

Does this pull request potentially affect one of the following parts:

• Dependencies (does it add or upgrade a dependency): (yes)
• Anything that affects deployment: (no)
• The MGS(Microservice Governance Services), i.e., Spring Cloud Gateway, OpenFeign, Eureka.: (no)

Documentation

• Does this pull request introduce a new feature? (no)

[peacewong]

LGTM.