Expose list membership with SQL null semantics in Java and Python

[robert3005]

Summary

Expose the membership expressions from #10052 in Java/JNI and Python, so callers can construct one set-membership expression with either Vortex or SQL null semantics.

Stacked on #10052, which builds on the SQL null semantics in #10057. DuckDB integration follows in #10054.

Changes

• Add Java listContains, in, and notIn, plus nonempty, empty, and null list literals and their JNI implementations.
• Add Python list_contains(..., sql_null_semantics=True) and in_list(value, list); negate in_list for SQL NOT IN.
• Fix Python typed-list construction and infer nullable list element types regardless of null position. Update exports, type stubs, and API documentation.
• Add Java scan and literal-construction coverage, and Python tests for null semantics, empty/null lists, serialization, scan filters, and large sets.

Validation: cargo clippy -p vortex-jni -p vortex-python --all-targets --all-features -- -D warnings, uvx ty check vortex-python, and Ruff on the changed expression stub passed. Java main and test sources compiled with ./gradlew :vortex-jni:compileJava :vortex-jni:compileTestJava -x buildJniLibrary -x makeTestFiles --console=plain (native compilation checked separately with Cargo). Runtime tests and benchmarks were not executed.

[codspeed]

Merging this PR will degrade performance by 11.19%

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

⚠️ 1 benchmark measured no execution time

Nothing ran under measurement, usually because the compiler removed the code under test. This result is not comparable, so it counts as unchanged.

Preventing compiler optimizations

• bench_compare_primitive[(10000, 2)]

⚠️ 12 benchmarks spent significant time in system calls

System calls cannot be consistently instrumented, so they are not included in the measure, which understates the real cost. Please switch to the Walltime instrument to accurately measure system calls.

Measurement and system calls

• from_vec_drop_vortex[65536]
• chunked_dict_primitive_canonical_into[u64, (1000, 1000, 10)]
• take_varbin
• multipolygon_mixed_validity
• zstd_compress_string
• ilike_contains
• pcodec_compress_f64
• decompress[u32, (1000, 256)]
• filter_run_end[len=4096_run=16_density=0.1]
• decompress[u16, (4000, 256)]
• …and 2 more

❌ 4 regressed benchmarks
✅ 2179 untouched benchmarks
⏩ 359 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	WallTime	compare_int_constant_neon	3.8 µs	4.4 µs	-12.09%
❌	WallTime	bitpack_blocked_compress_avx2	6.7 µs	7.6 µs	-11.39%
❌	WallTime	compare_u8_neon	1.9 µs	2.1 µs	-11.27%
❌	WallTime	deferred_bool_neon[i32, PerRowPerRow]	4.2 µs	4.7 µs	-10.02%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing rk/list-contains-bindings (446956a) with rk/list-contains-dict-runend (520e97b)}

Footnotes

1. 359 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩