chore(deps): allow better-sqlite3 13 as a peer (0.5.1)

[rodrigobnogueira]

Isolated major-version review: better-sqlite3 13

Widens the published peer range to ^11 || ^12 || ^13, matching nest-native/jobs#16 (released as 0.2.1). Both were needed: nest-native/reference-app#69 cannot install better-sqlite3 13 while either package's peer range rejects it (npm ERESOLVE).

What v13 actually changes

An N-API rewrite (prebuilt binaries ship with the package; prebuild-install is gone). The JavaScript surface is purely additive — db.explain() and statement.toString() added, nothing removed — plus a SqliteError cross-realm fix. Critically, SqliteError still carries the code the inbox's exactly-once dedup keys on: verified SQLITE_CONSTRAINT_UNIQUE on 13.0.2.

Verification (against 13.0.2, before widening)

• Full suite: 135/135 pass · Coverage: 100% statements/branches/functions/lines
• typecheck clean · release:check clean
• Sample smokes: showcase (atomic outbox → in-process dispatch → exactly-once inbox) and Kafka (outbox → KafkaOutboxTransport → @KafkaConsumer → exactly-once inbox) both pass

Why widened, not moved

better-sqlite3 13 requires Node >=22; this package supports Node >=20. The devDependency therefore stays on 12.x — otherwise the Node-20 CI legs stop installing and a supported Node is silently dropped. A new CI leg (better-sqlite3 13 compatibility (Node 22)) installs 13 and runs the suite there, so both ends of the range are tested. The guidelines now record that rule.

This also means #29 (bumping the better-sqlite3 devDependency to 13) should be closed rather than merged — it would break the Node-20 legs.

Behaviorally unchanged, so a patch release: 0.5.1.

[github-actions]

📊 Coverage Report

Metric		PR	Base	Diff
Statements	████████████████████	2070/2070 (100%)	2070/2070 (100%)	⚪ 0%
Branches	████████████████████	279/279 (100%)	279/279 (100%)	⚪ 0%
Functions	████████████████████	88/88 (100%)	88/88 (100%)	⚪ 0%
Lines	████████████████████	2070/2070 (100%)	2070/2070 (100%)	⚪ 0%

---

_{Updated for 149d3e1 | Compared against base branch}

[github-actions]

⏱️ Performance Report

✅ Tests	135 passed, 0 failed, 0 skipped
🧪 Suites	32
⏱️ Total step time	263.00s 🟢 -1.00s
⚙️ Test execution	255.34s 🟢 -867ms

🐢 Slowest test suites

#	Suite	Tests	Duration	vs Base
1	PostgresWakeListener	17	29.32s	🟢 -373ms
2	runWorkerLoop	13	23.26s	🔴 +561ms
3	OutboxClaimer (publish outcomes)	9	16.08s	⚪ ~0
4	WakeSocketServer + WakeSocketClient	9	15.84s	🔴 +276ms
5	KafkaInboxConsumer	8	13.89s	⚪ ~0
6	SqliteOutboxStore	8	13.79s	🟢 -184ms
7	OutboxWaker	7	12.73s	🔴 +712ms
8	MysqlOutboxStore	7	12.10s	⚪ ~0
9	InProcessOutboxTransport	5	8.79s	🔴 +117ms
10	SqliteInboxStore	5	8.48s	🟢 -276ms

🐌 Slowest individual tests

#	Suite	Test	Duration	vs Base
1	PostgresOutboxStore wakeChannel	PostgresOutboxStore wakeChannel enqueue fires pg_notify on the configured channel (observed via LISTEN)	3.42s	🟢 -230ms
2	OutboxProducer (atomic enqueue)	OutboxProducer (atomic enqueue) enqueue commits the outbox row with the business write	2.67s	⚪ ~0
3	OutboxProducer (atomic enqueue)	OutboxProducer (atomic enqueue) a throw rolls back BOTH the outbox row and the business write	2.06s	⚪ ~0
4	OutboxClaimer (publish outcomes)	OutboxClaimer (publish outcomes) tick publishes a pending row and marks it completed	1.95s	🟢 -96ms
5	WakeSocketServer + WakeSocketClient	WakeSocketServer + WakeSocketClient a non-EADDRINUSE bind error is rethrown as-is	1.94s	🔴 +279ms
6	PostgresWakeListener	PostgresWakeListener quotes a custom mixed-case channel (LISTEN must stay case-sensitive)	1.91s	🔴 +182ms
7	SqliteOutboxStore	SqliteOutboxStore claimBatch does NOT reclaim a freshly claimed processing row	1.90s	🔴 +130ms
8	transport errors	transport errors RetryableError carries name and optional delay	1.90s	🔴 +102ms
9	KafkaInboxConsumer	KafkaInboxConsumer reads a Buffer message key	1.89s	🔴 +316ms
10	MysqlOutboxStore	MysqlOutboxStore claimBatch marks candidates processing and returns the re-read rows	1.88s	🔴 +61ms
11	sqlite schema	sqlite schema inbox_events: composite (source, message_key) unique index	1.88s	🔴 +87ms
12	OutboxWaker	OutboxWaker notify() with a signal present still cleans up the abort listener	1.88s	🔴 +258ms
13	WakeSocketServer + WakeSocketClient	WakeSocketServer + WakeSocketClient client notify() without onError still swallows the failure	1.87s	⚪ ~0
14	runWorkerLoop	runWorkerLoop a throwing tick without onError still continues (onError is optional)	1.87s	⚪ ~0
15	PostgresOutboxStore	PostgresOutboxStore markCompleted / retry / markFailed transition the row	1.87s	⚪ ~0

---

_{Updated for 149d3e1 | Compared against base branch}

[github-actions]

🧠 Cognitive Complexity Report

Metric	PR	Base	Diff
Total complexity	79	79	⚪ 0
Max function complexity	12	12	⚪ 0
Functions measured	39	39	⚪ 0

🧩 Most complex functions

Function	Location	Complexity	Status
runWorkerLoop	packages/messaging/outbox-worker.ts:34	12	🟢 OK
KafkaInboxConsumer.consume	packages/messaging/adapters/kafka/kafka-inbox-consumer.ts:67	5	🟢 OK
PostgresWakeListener.#supervise	packages/messaging/dialects/postgres/wake.ts:122	5	🟢 OK
WakeSocketServer.listen	packages/messaging/wake-socket.ts:52	5	🟢 OK
OutboxClaimer.onPublishError	packages/messaging/outbox-claimer.service.ts:84	4	🟢 OK
deriveDedupKey	packages/messaging/wire-contract.ts:55	4	🟢 OK
MysqlInboxStore.runOnce	packages/messaging/dialects/mysql/inbox-store.ts:43	3	🟢 OK
PostgresInboxStore.runOnce	packages/messaging/dialects/postgres/inbox-store.ts:37	3	🟢 OK
SqliteInboxStore.runOnce	packages/messaging/dialects/sqlite/inbox-store.ts:39	3	🟢 OK
KafkaInboxConsumer.readKey	packages/messaging/adapters/kafka/kafka-inbox-consumer.ts:90	2	🟢 OK

🧭 Cognitive complexity is reported as a review signal, not a merge gate. Prefer small, intention-revealing refactors when complexity rises.

---

_{Updated for 149d3e1 | Compared against base branch}