fix(server): handle repeated range predicates correctly

[LegendPei]

Purpose of the PR

• close [Bug] Wrong result for repeated predicates #3083

Fix incorrect empty results when multiple predicates target the same range-indexed property.

Main Changes

• Cache the range-index validation result per element and property.
• Avoid consuming the same index value multiple times for repeated predicates.
• Add regression coverage for repeated predicates, including reversed predicate order.

Verifying these changes

• Trivial rework / code cleanup without any test coverage. (No Need)
• Already covered by existing tests, such as (please modify tests here).
• Need tests and can be verified as follows:
  • mvn test -pl hugegraph-server/hugegraph-test -am -P core-test,rocksdb -Dtest=EdgeCoreTest -DfailIfNoTests=false

Does this PR potentially affect the following parts?

• Dependencies (add/update license info & regenerate_known_dependencies.sh)
• Modify configurations
• The public API
• Other affects (typed here)
• Nope

Documentation Status

• Doc - TODO
• Doc - Done
• Doc - No Need

[imbajin]

Blocking: yes. Summary: The query-lifetime range-match cache reuses an incomplete first-batch result and causes valid elements to be omitted. Evidence: three independent review lanes, exact-head static tracing, and repeated failures of VertexCoreTest.testRemoveLeftRangeIndex across memory, RocksDB, HBase, macOS, and HStore checks.

[imbajin]

Blocking: yes. Summary: The correctness fix introduces an allocation on every condition-filtered element, including queries that never use range-index bookkeeping. Evidence: exact-head static inspection and targeted memory-profile path analysis; the two focused correctness tests pass.

[Copilot]

Pull request overview

This PR fixes incorrect empty results when multiple has() predicates target the same range-indexed property (issue #3083), by preventing range-index validation from “consuming” the same index value multiple times during per-element filtering in ConditionQuery.

Changes:

• Cache range-index validation results per element + property while evaluating conditions, so repeated predicates on the same range-indexed property don’t invalidate otherwise-matching elements.
• Adjust Element2IndexValueMap.checkRangeIndex() to accept a per-element cache map and reuse previously computed match outcomes for the same property.
• Add a regression test covering repeated predicates and reversed predicate order.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
hugegraph-server/hugegraph-test/src/main/java/org/apache/hugegraph/core/EdgeCoreTest.java	Adds a regression test for repeated range predicates on the same edge property.
hugegraph-server/hugegraph-core/src/main/java/org/apache/hugegraph/backend/query/ConditionQuery.java	Introduces per-element caching for range-index validation to avoid double-consuming index values for repeated predicates.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[imbajin]

Blocking: no. Summary: The cross-batch correctness and non-index allocation regressions are fixed; the remaining range-index allocation concern is already covered by an existing thread, so no duplicate inline comment is added. Evidence: three independent exact-head review lanes, EdgeCoreTest ran 169 tests with 0 failures/errors and 11 skipped, and visible exact-head checks are green.