feat(queen): enforce the measured runtime worker limit

[gHashTag]

Closes browseros-ai#1311 and implements Wave C of browseros-ai#1321.

Observable contract

The server measures one effective worker limit for the round and sends that same numeric limit to every queend choose call. The compiled policy clamps explicit limits to 1...8 and keeps the legacy default of 4 when older callers omit the field.

Change

• derive the round limit from distinct configured worker capacity
• freeze it for the complete round, including refill choices
• enforce it in the compiled Swift policy and report the effective refusal limit
• keep the app and Linux policy mirrors byte-identical
• make the operator board read the same measured capacity and limit

Verification

• RED reproduced: an explicit limit of 8 was ignored as limit 4
• TRIOS_REQUIRE_QUEEND_TESTS=1 bun test apps/server/tests/api/queend-choose.test.ts — 18 passed
• TRIOS_REQUIRE_QUEEND_TESTS=1 bun test apps/server/tests/api/queen-round.test.ts — 31 passed
• full Queen API suite — 156 passed, 0 failed
• standalone Swift delegation suite — 137 passed, 0 failed
• bun run --filter @browseros/server typecheck — passed
• policy mirror cmp — identical
• lefthook pre-commit/pre-push — passed

[github-actions]

❌ Tests failed — 3/1321 failed

Suite	Passed	Failed	Skipped
✅ agent	80/80	0	0
✅ build	9/9	0	0
✅ cdp-protocol	5/5	0	0
✅ eval	93/93	0	0
✅ server-agent	262/262	0	0
❌ server-api	440/476	2	34
✅ server-browser	4/4	0	0
✅ server-integration	10/11	0	1
✅ server-lib	271/271	0	0
✅ server-root	64/64	0	0
✅ server-skills	31/31	0	0
❌ server-tools	0/1	1	0
✅ shared	14/14	0	0

Failed tests

• server-api — the migration block, applied to a real PostgreSQL > creates every object it promises, and survives a second boot
• server-api — the live gate > ran, or its absence is on the record
• server-tools — workflow > server-tools setup

View workflow run