fix(query): preserve count and predicate semantics (2/3)

[contrueCT]

Purpose of the PR

Part 2 of a three-PR upgrade series targeting apache/hugegraph:master. Preserve transaction count and predicate semantics while retaining TinkerPop 3.5.1.

Merge order: apache#3261 (Java 17 foundation) → apache#3262 (query semantics) → apache#3263 (TinkerPop upgrade). All three PRs target master directly. Merge this PR after apache#3261. Until the prerequisite is merged, the Files changed tab includes its changes; the part-2-only diff is available in community #232.

Source: hugegraph/hugegraph#232, submitted directly from hugegraph:task/tp381-2-query-semantics. The local validation statements below are carried over from that source PR; CI on this ASF PR head remains the merge gate.

Relationship to other work

• This three-part series is the split delivery path for the modernization tracked in apache#3117 and issue #3069. BREAKING CHANGE(server): upgrade Java17 + TP3.7 + Groovy4 #3117's current code also targets TinkerPop 3.8.1; the series is intended to replace that monolithic delivery, rather than add another independent upgrade.

• Generic label/ID/SEARCH condition resolution and selective predicate pushdown remain with apache#2994, issue #3201 and candidate-index coverage in apache#3243. This series retains whole-step local filtering and transaction/count fixes; it adds no generalized partial extraction or optimizer-time index-coverage heuristic. Existing special handling around match() and connective label filters is inherited from master.

• community #260 relocates shared query/model types, and community #261 changes transaction lifecycle. Their overlapping engine files need integration coordination; those migrations are not folded into this series.

• Single-ID query fast paths (apache#3175, apache#2859) and adjacency-query optimization (apache#2864) remain separate performance work. Cypher parameter binding (community #238) and error mapping (apache#3259, apache#3241) remain separate behavior changes; part 3 only adapts upgraded transport APIs and result normalization.

• The Java 17 launcher checks also cover the older Java 11 startup-check goal in apache#2846. Configurable distribution paths (apache#3253) and restart diagnostics (apache#3258) touch the same scripts and remain separate integration work.

Main Changes

• Count uncommitted vertex/edge changes through the transaction query path and always close fallback iterators. Continue rejecting limit, offset, paging and unsupported aggregates with uncommitted changes.
• For ordinary GraphStep/VertexStep extraction, keep the whole HasStep local when any sibling condition is unsupported. Retain the existing special handling around match() and connective label filters from master. Count optimization does not skip filtering barriers.
• Reset optimized count execution state and exclude result iterators from query-step equality. This fixes the existing count equality regression.

TinkerPop-specific NotP, serializer and step-API changes remain in part 3. This PR still compiles and runs against TinkerPop 3.5.1.

Verifying these changes

• Already covered by existing tests: CountStrategyCoreTest, TraversalUtilOptimizeTest, GraphTransactionTest, and QueryListTest.
• Local validation: Formatting and all-module clean compile passed. 82 focused core/optimizer tests passed on each of Memory and RocksDB against TinkerPop 3.5.1, plus 17 GraphTransaction/QueryList unit tests in their separate unit-test profile.

Does this PR potentially affect the following parts?

• Dependencies
• Modify configurations
• The public API
• Other affects: transaction count and traversal optimization behavior
• Nope

Documentation Status

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

Documentation: docs/query-semantics.md.

Review follow-up

• Pending removals take precedence over added/updated records, so update-then-remove yields zero in both lists and counts, including self-loops. Same-ID re-adds remain visible.
• Document the existing dirty-index boundary: conditions using an index reject uncommitted index changes; native label scans, global counts and direct-ID queries retain their supported behavior. Real Memory and RocksDB regressions cover these paths.
• Immutable label collections such as List.of(...) and Set.of(...) are inspected by iteration; predicates that actually contain null labels retain local filtering. Optimizer and query regressions cover both cases.
• Unsupported mixed native ID/text predicates stay together in the local HasStep. Regression checks cover matching and nonmatching results, counts, pagination and plan retention with the HugeGraph strategies enabled and disabled.
• Preserve transaction-visible counts, self-loop multiplicity, backend system counts and filtering barriers. Ordinary ID equality remains covered.
• Inherit the Java 11 bytecode target for published Commons/RPC libraries and Ivy 2.6.0 from the foundation; build and service runtimes remain Java 17.
• Java 17 formatting and all-module clean compile pass. The CountStrategy/optimizer regressions pass against RocksDB; runtime Grape loading is validated in the foundation.

CI on the updated PR head remains separate verification.

[codecov]

Codecov Report

❌ Patch coverage is 85.96491% with 16 lines in your changes missing coverage. Please review.
✅ Project coverage is 37.15%. Comparing base (0a3e4ae) to head (1a53ed4).

Files with missing lines	Patch %	Lines
...rg/apache/hugegraph/store/meta/GraphIdManager.java	85.96%	8 Missing and 8 partials ⚠️

Additional details and impacted files

@@             Coverage Diff              @@
##             master    #3262      +/-   ##
============================================
- Coverage     41.60%   37.15%   -4.46%     
+ Complexity     7320     1741    -5579     
============================================
  Files           794      255     -539     
  Lines         69127    17546   -51581     
  Branches       9258     1894    -7364     
============================================
- Hits          28761     6519   -22242     
+ Misses        37091    10495   -26596     
+ Partials       3275      532    -2743     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.