release: 0.0.10 — graph perf, identity map, postgres joins

[eldonm]

Summary

Release 0.0.10. Graph-perf substrate consumed by the downstream integral perf audit.

• Request-scoped identity map (dedupes node hydration within a unit of work)
• Single-hop find_connected_nodes traversal primitive
• Auth service reuse (fewer redundant auth resolutions)
• Walker batching knobs
• Observable db_op coverage for query instrumentation
• Postgres join support

Prepares the dependency surface for integral's perf work (identity map + find_connected_nodes + query budgets).

Test plan

• CI green (existing suite)
• Downstream: integral feat/jv-0.0.10-perf-upgrade pins jvspatial==0.0.10 and its query-budget guards pass

🤖 Generated with Claude Code

[github-actions]

Benchmark comparison

Threshold: ±25% (informational, does not block merge)

benchmark	baseline (s)	current (s)	delta	status
tests/benchmarks/test_deferred_save_benchmarks.py::test_bench_deferred_save_batched_100	0.018735	0.039068	+108.5%	REGRESSION (+108.5%)
tests/benchmarks/test_deferred_save_benchmarks.py::test_bench_immediate_save_100	0.019049	0.038359	+101.4%	REGRESSION (+101.4%)
tests/benchmarks/test_jsondb_benchmarks.py::test_bench_jsondb_batched_saves_500	0.667799	0.403037	-39.6%	IMPROVED (-39.6%)
tests/benchmarks/test_jsondb_benchmarks.py::test_bench_jsondb_count_empty_query	1.241775	0.889170	-28.4%	IMPROVED (-28.4%)
tests/benchmarks/test_jsondb_benchmarks.py::test_bench_jsondb_count_filtered	0.761262	1.015276	+33.4%	REGRESSION (+33.4%)
tests/benchmarks/test_jsondb_benchmarks.py::test_bench_jsondb_find_filtered	0.919747	0.831270	-9.6%	OK
tests/benchmarks/test_jsondb_benchmarks.py::test_bench_jsondb_save_throughput	0.005320	0.001526	-71.3%	IMPROVED (-71.3%)
tests/benchmarks/test_sqlite_benchmarks.py::test_bench_sqlite_count_empty	0.122752	0.234518	+91.1%	REGRESSION (+91.1%)
tests/benchmarks/test_sqlite_benchmarks.py::test_bench_sqlite_count_fallback_via_regex	0.149793	0.315153	+110.4%	REGRESSION (+110.4%)
tests/benchmarks/test_sqlite_benchmarks.py::test_bench_sqlite_count_pushdown	0.132947	0.250449	+88.4%	REGRESSION (+88.4%)
tests/benchmarks/test_sqlite_benchmarks.py::test_bench_sqlite_find_fallback_via_regex	0.143852	0.299285	+108.1%	REGRESSION (+108.1%)
tests/benchmarks/test_sqlite_benchmarks.py::test_bench_sqlite_find_pushdown	0.124463	0.247695	+99.0%	REGRESSION (+99.0%)
tests/benchmarks/test_sqlite_benchmarks.py::test_bench_sqlite_sort_limit_pushdown	0.145788	0.277250	+90.2%	REGRESSION (+90.2%)