feat: 30.2 — external sdk unification (5 sdks)

[vieiralucas]

Summary

• All 5 external SDKs updated to unified API surface from Story 30.1
• PRs opened in each SDK repo (parallel agents, ~6-10 min each):
  • Go: feat: 30.2 — unified api surface fila-go#4
  • Python: feat: 30.2 — unified api surface fila-python#4
  • JS: feat: 30.2 — unified api surface fila-js#4
  • Ruby: feat: 30.2 — unified api surface fila-ruby#4
  • Java: feat: 30.2 — unified api surface fila-java#4

Changes per SDK

• batch_enqueue → enqueue_many (or language-idiomatic equivalent)
• BatchMode → AccumulatorMode
• Ack/Nack wrap single items in repeated fields
• Consume only reads messages (no singular fallback)
• No "batch" prefix in any public API

Test plan

• Go: 20/20 tests pass
• Python: 31/31 tests pass
• JS: 30/30 tests pass
• Ruby: tests pass
• Java: 34/35 pass (1 pre-existing TLS failure on main)

🤖 Generated with Claude Code

---

Summary by cubic

Unified all 5 external SDKs (Go, Python, JS, Ruby, Java) to the Story 30.1 API surface, removing batch-prefixed APIs and standardizing enqueue/consume/ack/nack shapes. Updated planning artifacts to mark Story 30.2 as done; CI is green across SDKs except one known Java TLS flake.

• Migration
  • Replace batch_enqueue with enqueue_many (language-idiomatic names).
  • Rename BatchMode to AccumulatorMode.
  • enqueue accepts one or more messages; consume reads messages only.
  • ack/nack accept one or more message IDs.
  • Remove any batch prefix/suffix from public APIs.

^{Written for commit 68cf818. Summary will update on new commits.}

Benchmark Results (vs main baseline)

Baseline commit: 93584a6 PR commit: c216c8d Threshold: 10%

Benchmark	Baseline	Current	Change	Unit
compaction_active_enqueue_max	41.18	41.28	+0.2%	ms
compaction_active_enqueue_p50	0.66	0.68	+3.2%	ms
compaction_active_enqueue_p95	0.70	0.72	+2.8%	ms
compaction_active_enqueue_p99	0.78	0.83	+7.5%	ms
compaction_active_enqueue_p99_9	40.77	40.80	+0.1%	ms
compaction_active_enqueue_p99_99	41.15	41.22	+0.2%	ms
compaction_idle_enqueue_max	41.22	41.89	+1.6%	ms
compaction_idle_enqueue_p50	0.31	0.32	+3.8%	ms
compaction_idle_enqueue_p95	0.35	0.36	+3.2%	ms
compaction_idle_enqueue_p99	0.37	0.40	+5.9%	ms
compaction_idle_enqueue_p99_9	40.73	40.86	+0.3%	ms
compaction_idle_enqueue_p99_99	41.15	41.22	+0.2%	ms
compaction_p99_delta	0.38	0.45	+19.5%	ms	🔴
consumer_concurrency_100_throughput	1782.00	1662.67	-6.7%	msg/s
consumer_concurrency_10_throughput	1119.00	1083.33	-3.2%	msg/s
consumer_concurrency_1_throughput	130.33	103.33	-20.7%	msg/s	🔴
e2e_latency_light_max	42.53	42.40	-0.3%	ms
e2e_latency_light_p50	40.54	40.58	+0.1%	ms
e2e_latency_light_p95	40.70	41.41	+1.7%	ms
e2e_latency_light_p99	41.41	41.47	+0.2%	ms
e2e_latency_light_p99_9	41.50	41.57	+0.2%	ms
e2e_latency_light_p99_99	42.53	42.40	-0.3%	ms
enqueue_throughput_1kb	3486.96	2936.88	-15.8%	msg/s	🔴
enqueue_throughput_1kb_mbps	3.41	2.87	-15.8%	MB/s	🔴
equal_weight_fairness_jains_index	1.00	1.00	+0.0%	index
equal_weight_fairness_max_deviation	0.00	0.00	n/a	% deviation
equal_weight_fairness_tenant-1	0.00	0.00	n/a	% deviation
equal_weight_fairness_tenant-2	0.00	0.00	n/a	% deviation
equal_weight_fairness_tenant-3	0.00	0.00	n/a	% deviation
equal_weight_fairness_tenant-4	0.00	0.00	n/a	% deviation
equal_weight_fairness_tenant-5	0.00	0.00	n/a	% deviation
fairness_accuracy_jains_index	1.00	1.00	+0.0%	index
fairness_accuracy_max_deviation	0.20	0.20	+0.0%	% deviation
fairness_accuracy_tenant-1	0.20	0.20	+0.0%	% deviation
fairness_accuracy_tenant-2	0.20	0.20	+0.0%	% deviation
fairness_accuracy_tenant-3	0.10	0.10	+0.0%	% deviation
fairness_accuracy_tenant-4	0.10	0.10	+0.0%	% deviation
fairness_accuracy_tenant-5	0.10	0.10	+0.0%	% deviation
fairness_overhead_fair_throughput	1103.32	1037.16	-6.0%	msg/s
fairness_overhead_fifo_throughput	1106.24	1068.43	-3.4%	msg/s
fairness_overhead_pct	0.29	3.08	+973.3%	%	🔴
key_cardinality_10_throughput	1273.29	1280.27	+0.5%	msg/s
key_cardinality_10k_throughput	506.56	503.62	-0.6%	msg/s
key_cardinality_1k_throughput	778.65	759.60	-2.4%	msg/s
lua_on_enqueue_overhead_us	27.41	36.11	+31.7%	us	🔴
lua_throughput_with_hook	874.09	868.55	-0.6%	msg/s
memory_per_message_overhead	169.16	1794.05	+960.5%	bytes/msg	🔴
memory_rss_idle	427.90	407.88	-4.7%	MB
memory_rss_loaded_10k	429.99	425.02	-1.2%	MB

Summary: 7 regressed, 0 improved, 42 unchanged

⚠️ Performance regression detected — 7 metric(s) exceeded the 10% threshold

[cubic-dev-ai]

No issues found across 3 files