feat(gq): add list membership with scan pushdown

[azimafroozeh]

What & why

GQ rejects $m.number in $numbers. Add in for read filters, projections and mutation predicates, with scan pushdown for literal/parameter lists and bounded row estimates for membership on key or unique properties.

Handle empty lists, null elements, Date constants and mixed numeric literals. Allow quoted traversal names such as $a "in" $b so reserving in does not make the edge inaccessible. Add row and plan coverage for traversal access paths, nearest ranking, and indexed data across writes, deletion and restart.

Backing issue / RFC

Closes #801.

Checklist

• Change is focused on list membership and its execution paths.
• Tests added/updated for behavior changes.
• Public query/mutation docs and skill references updated.
• Reviewed against docs/dev/invariants.md.

Local verification

• cargo test -p omnigraph-gqt --locked --test gq_logic_tests: 220 passed.
• cargo test -p omnigraph-gqt --locked --lib --test runner_dispatch: 183 library tests and 12 dispatch tests passed.
• cargo test -p omnigraph-engine --locked --test lance_surface_guards scalar_index_use_requires_matched_literal_type: passed, including four-value membership index checks.
• cargo test --workspace --exclude omnigraph-gqt --exclude omnigraph-dst --locked --features omnigraph-engine/failpoints,omnigraph-cluster/failpoints --no-fail-fast: 120 test binaries passed; four known sandbox failures remain (three CLI managed::data tests and external_blob_file_policy_rejects_special_files).
• cargo fmt --all --check and cargo clippy --workspace --all-targets --locked -- -D warnings -W clippy::dbg_macro: passed.
• python3 scripts/check-docs.py and bash scripts/check-agents-md.sh: passed.

Notes for reviewers

Pushed numeric membership follows existing = coercion. This also changes constant-list contains from its previous in-memory behavior; the existing numeric precision difference between scan filters and projections remains outside this change.

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, add credits to your account and enable them for code reviews in your settings.