test(qa-callers): deterministic child transport, completion-time handoff reads, compact phase report

The exploratory caller cases exist to prove the caller starts and bounds
exploratory QA. Their native adversarial reviewer (review) and plan audit
(ship plan-checks) now come from recorded child outputs instead of a live
subagent, handoff freshness reads are required before completion records
rather than every bookkeeping log, and the phase report is compact. Measured:
194-257 s per case against 208-284 s before, no subagent calls.

test/qa-exploratory-callers.test.ts

@@ -1365,7 +1365,7 @@ describe('real caller-specific native fixture and capture boundary', () => {
         expect(options.appendSystemPrompt).not.toMatch(/bun scripts\/probe\.ts \d|invalid input|highest.risk/i);
         expect(options.prompt).toContain('Keep normal parent decision gates.');
         expect(options.prompt).toContain("use the section clock's Hard deadline UTC, never its Runner entry UTC, reserve-start time or a clock-read time");
-        expect(options.prompt).toContain('Before every completion report or bookkeeping log, read HANDOFF.md');
+        expect(options.prompt).toContain('Before the completion report and each completed:true review record, read HANDOFF.md');
         expect(options.prompt).toContain('a later handoff read cannot validate an earlier completion');
         expect(options.prompt).toContain('If you defer an optional idea or stop exploration, do not publish a checkpoint for it');
         expect(options.prompt).toContain('An unused checkpoint requires an actual authenticated expired-capture result; nearing the deadline or choosing to stop is not enough');