BitSetConjunctionDISI iterators fall out of sync when lead doc >= minlength of bitsets [LUCENE-10674]

In [BitSetConjunctionDISI.doNext()](https://github.com/apache/lucene/blob/releases/lucene/9.3.0/lucene/core/src/java/org/apache/lucene/search/ConjunctionDISI.java#L283-L285) if the lead doc is greater than or equal to the length of any of the BitSets, NO_MORE_DOCS is returned. On subsequent calls to [BitSetConjunctionDISI.docId(),](https://github.com/apache/lucene/blob/releases/lucene/9.3.0/lucene/core/src/java/org/apache/lucene/search/ConjunctionDISI.java#L263) the lead's docID which is not exhausted will be returned. I think this could be fixed by calling lead.advance(NO_MORE_DOCS) before returning NO_MORE_DOCS in doNext.

Related issue: #10581

---
Migrated from [LUCENE-10674](https://issues.apache.org/jira/browse/LUCENE-10674) by Jack Mazanec, updated Aug 15 2022
Pull requests: https://github.com/apache/lucene/pull/1068

Start with BitSetConjunctionDISI.doNext() and docId() in lucene/core/src/java/org/apache/lucene/search/ConjunctionDISI.java, using the linked issue #10581 for context. Reproduce the case where the lead doc is greater than or equal to a BitSet length, then verify that subsequent docId() calls consistently report exhaustion. The issue body references pull request 1068, so check that work before proceeding.

Written by the indexing model from the issue text.