Batch segment retrieval from the metadata store

[abhishekrb19]

Used segments retrieval fails when there are too many intervals (about 120 with Derby). Same thing can happen with multiple intervals as with unused segments.

Previously, the intervals weren't capped in the SQL query and is bloated by the fact that we add grouped OR clause per interval to handle the eternity interval here. This PR splits up the SELECT query into multiple batches, with 100 intervals/batch. This is similar to the batching strategy with a cap on max number of segments announced at once.

org.skife.jdbi.v2.exceptions.CallbackFailedException: org.skife.jdbi.v2.exceptions.UnableToCreateStatementException: 

java.sql.SQLSyntaxErrorException: Statement too complex. Try rewriting the query to remove complexity. Eliminating many duplicate expressions or breaking up the query and storing interim results in a temporary table can often help resolve this error. 
[statement:"SELECT payload FROM foo WHERE used = :used AND dataSource = :dataSource AND ((start < :end0 AND \"end\" > :start0) OR (start = '-146136543-09-08T08:23:32.096Z' AND "end" != '146140482-04-24T15:36:27.903Z' AND "end" > :start0) OR (start != '-146136543-09-08T08:23:32.096Z' AND "end" = '146140482-04-24T15:36:27.903Z' AND start < :end0) OR (start < :end1 AND \"end\" > :start1) OR (start = '-146136543-09-08T08:23:32.096Z' AND "end" != '146140482-04-24T15:36:27.903Z' AND "end" > :start1) OR (start != '-146136543-09-08T08:23:32.096Z' AND "end" = '146140482-04-24T15:36:27.903Z' AND start < :end1) OR (start < :end2 AND \"end\"

Core changes:

• Segment retrieval is batched at the storage layer by intervals. It's capped at 100 intervals per batch.
• With multiple batches, the iterators are concatenated together and returned to the caller.
• Add a combination of unit tests that test the different scenarios with limit in, at and out of range scenarios with used and unused segments retrieval.

A fix localized to kill tasks that originally exposed this bug is available here - #15306. We'd still separately want this change in the server as the issue can happen more broadly.

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.

[zachjsh]

LGTM