Adding new replica lag metric

[Pankaj260100]

Description

• Adding a new replica lag metric along with taskId, datasource, and stream dimensions.

Let's say the offsets of a task group with 3 partitions are:
replica0 : {0: 100, 1: 200, 3: 300}
replica1: {0: 120, 1: 180, 3: 250}
The lag for replica0 is (120 - 100) + 0  + 0 = 20
The lag for replica1 is 0 + (200 - 180) + (300 - 250) = 70

Fixed the bug ...

Renamed the class ...

Added a forbidden-apis entry ...

Release note

---

Key changed/added classes in this PR

• MyFoo
• OurBar
• TheirBaz

---

This PR has:

• been self-reviewed.
  • using the concurrency checklist (Remove this item if the PR doesn't have any relation to concurrency.)
• added documentation for new or modified features or behaviors.
• a release note entry in the PR description.
• added Javadocs for most classes and all non-trivial methods. Linked related entities via Javadoc links.
• added or updated version, license, or notice information in licenses.yaml
• added comments explaining the "why" and the intent of the code wherever would not be obvious for an unfamiliar reader.
• added unit tests or modified existing tests to cover new code paths, ensuring the threshold for code coverage is met.
• added integration tests.
• been tested in a test Druid cluster.

[AmatyaAvadhanula]

@Pankaj260100

1. This metric might not be of interest for every user. Especially those operating with exactly 1 replica. Could you please add a config in the supervisor to allow or disable emission of this metric? (Alternatively, emit this only when there are multiple replicas)
2. Please add relevant documentation and modify the PR description appropriately.
3. It would be great if you could try adding this metric for all the streaming supervisors as a generic method.

[cryptoe]

This metric might not be of interest for every user. Especially those operating with exactly 1 replica. Could you please add a config in the supervisor to allow or disable emission of this metric? (Alternatively, emit this only when there are multiple replicas)

@AmatyaAvadhanula If you think this metric is useful, lets enable it for all. If its not useful, then we should not add this. Adding more context parameter to supervisor spec seems another turning knob which we want to avoid.