CNDB-12460: Fix PQVector reencoding when refining in CompactionGraph

[jkni]

What is the issue

When refining PQVectors in CompactionGraph, indexing of vectors may fail because vectors in the graph aren't yet encoded, or vectors may incorrectly be encoded as all zeroes.

What does this PR fix and why was it fixed

When re-encoding vectors after refinement, use the previous count of compressed vectors rather than the max ordinal in the graph. Because adding to the graph is asynchronous, there may be encoded vectors that aren't in the graph. The old approach would not re-encode any vectors with ordinals above the max ordinal. Depending on timing, this could cause vectors to fail to index (if they try to compare to an ordinal that isn't in the encoded vectors) or be encoded as all zeroes (due to the code that fills holes in compressed vectors).

Checklist before you submit for review

• Make sure there is a PR in the CNDB project updating the Converged Cassandra version
• Use NoSpamLogger for log lines that may appear frequently in the logs
• Verify test results on Butler
• Test coverage for new/modified code is > 80%
• Proper code formatting
• Proper title for each commit staring with the project-issue number, like CNDB-1234
• Each commit has a meaningful description
• Each commit is not very long and contains related changes
• Renames, moves and reformatting are in distinct commits

[jbellis]

Restating the problem:

Although compressedVectors.count() == builder.getGraph().getIdUpperBound() when all insertions have completed, these are not equivalent during construction since addGraphNode (which modifies the graph upper bound) is called separately and asynchronously, so refine() needs to use the former instead of the latter.

Fix LGTM.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
100.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[cassci-bot]

❌ Build ds-cassandra-pr-gate/PR-1506 rejected by Butler

---

1 new test failure(s) in 1 builds
See build details here

---

Found 1 new test failures

Test	Explanation	Branch history	Upstream history
...ToolEnableDisableBinaryTest.testMaybeChangeDocs	regression	🔴	🔵🔵🔵🔵🔵🔵🔵

---

Found 3 known test failures