refactor(java11) - convert most modules to java 11

[leifker]

This PR is one of two Java 11 options (other) - only merge one. This version converts modules to Java 11 where they are not tied to spark-lineage which requires Java 8. Unlike the other PR, there are no duplicate modules for Java 11 and Java 8, only one java version is chosen based on its usage by spark-lineage

All modules except these are converted to java 11:

li-utils
metadata-models
test-models
metadata-integration:java:datahub-client
metadata-integration:java:spark-lineage

Checklist

• The PR conforms to DataHub's Contributing Guideline (particularly Commit Message Format)
• Links to related issues (if applicable)
• Tests for the changes have been added/updated (if applicable)
• Docs related to the changes have been added/updated (if applicable). If a new feature has been added a Usage Guide has been added for the same.
• For any breaking change/potential downtime/deprecation/big changes an entry has been made in Updating DataHub

[github-actions]

Unit Test Results (metadata ingestion)

       8 files  ±0         8 suites  ±0   55m 36s ⏱️ - 2m 56s
   717 tests ±0     714 ✔️ ±0  3 💤 ±0  0 ❌ ±0 
1 436 runs  ±0  1 430 ✔️ ±0  6 💤 ±0  0 ❌ ±0 

Results for commit fb4d40c. ± Comparison against base commit 325b959.

♻️ This comment has been updated with latest results.

[github-actions]

Unit Test Results (build & test)

584 tests  ±0   580 ✔️ ±0   12m 54s ⏱️ -1s
143 suites ±0       4 💤 ±0 
143 files   ±0       0 ❌ ±0 

Results for commit fb4d40c. ± Comparison against base commit 325b959.

♻️ This comment has been updated with latest results.

[pedro93]

@pedro93 - The intent is to build any, or all, modules with jdk11. The docker images would need to updated to jdk11 as a requirement to run the jars generated with jdk11. Only certain modules would require jdk8, specifically those for supporting spark-lineage. For this later case, those specific modules can either be compile for jdk8 or for both jdk8 and jdk11 (this is the difference between the two PRs).

My question is more, what is the benefit of compiling datahub code with java 11 jdk? byte code improvements? Java 11 language features?

Following #5875 all datahub images will run in Java 11 runtime with java 8 byte code.

[leifker]

There are a couple benefits of compiling with jdk11 where possible.

One of them are language features for sure. For example, :metadata-integration:java:datahub-protobuf, in the project is using jdk11 and is not compatible with jdk8. I wrote that module before realizing the jdk8 dependence of the project.

I think that the most important reason is ultimately that many libraries are deprecating jdk8 support and without updating to jdk11 this locks DataHub into older versions of libraries. In turn, these older libraries may not contain important security fixes and newer functionality.

[shirshanka]

LGTM!