feat(rectification): range delivery needs 4 dated events across 3 domains (R1, BUG-1193)

Product decision 2026-10-02 (TASK-upstream-sync5 R1): a time range is offered
only with at least 4 dated, primary-scoreable events covering 3 domains,
counted on all of them (training + reserved holdout). Was 3 training events /
2 domains in three TS copies and the Python acceptance gate while the policy
file already said 4/3.

- One definition: references/rectification_policy.v1.json
  (minConfirmationEvents / minConfirmationDomains). TS core/types MIN_DATED_*,
  rectification-decision MIN_STANDALONE_*, evidence-model MIN_ACCEPTANCE_*,
  the convergence evaluator and the post-inference trainingGateOpen all read
  it; Python decision_policy MIN_ACCEPTANCE_* alias MIN_CONFIRMATION_*.
- Python receipt counts all scoreable events / domains for event_quality and
  domain_diversity; decision policy identity v3 -> v4 (candidate UUIDs carry
  it). Candidate scores unchanged (77 v5 cases A/B identical), so the
  algorithm stays rectification-v5-matrix-scoring-10.
- Memoization golden v3 written by write_golden; v2 frozen by sha256 with a
  test that its scores equal v3 and only the receipt policy moved.
- Collect gap copy names the exact gap ("再来两件……其中至少一件不是……")
  instead of always "再来一件"; VOICE.md updated. Legacy life-events form copy
  4/3 as well.
- 30 frontend test files, 4 Python tests: fixtures extended to the same
  scenario at 4/3, or assertions changed with 原值/新值/原因 notes.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01N4f2nya58RoRu4yEmJgRGE

scripts/rectification/decision_policy.py

@@ -26,13 +26,16 @@ from scripts.rectification_policy import (
     MIN_CONFIRMATION_MARGIN_PERCENT,
 )
 
-POLICY_VERSION = "rectification-candidate-policy-v3"
+POLICY_VERSION = "rectification-candidate-policy-v4"
 RECEIPT_VERSION = "candidate-decision-receipt-v2"
 EXECUTION_LEDGER_VERSION = "rectification-execution-ledger-v2"
 SCORE_QUANTUM = Decimal("0.0001")
 TIE_ABSOLUTE_TOLERANCE = Decimal("0.0001")
-MIN_ACCEPTANCE_EVENTS = 3
-MIN_ACCEPTANCE_DOMAINS = 2
+# R1 (2026-10-02, BUG-1193): a range is accepted only with at least 4 dated
+# scoreable events across 3 domains, counted on all of them (training + the
+# reserved holdout). Single definition: references/rectification_policy.v1.json.
+MIN_ACCEPTANCE_EVENTS = MIN_CONFIRMATION_EVENTS
+MIN_ACCEPTANCE_DOMAINS = MIN_CONFIRMATION_DOMAINS
 MIN_DATE_QUALITY_MEAN = Decimal("0.65")
 MIN_DIAGNOSTIC_RETENTION = Decimal("0.75")
 MIN_ACCEPTANCE_MARGIN_PERCENT = Decimal("10")
@@ -543,17 +546,19 @@ def build_decision_receipt(
         if str(event.get("id") or "") not in holdout
     ]
     domains = sorted({event["domain"] for event in training_events})
+    scoreable_domains = sorted({event["domain"] for event in scoreable_events})
     candidate_presence = _gate(bool(candidate_decisions), candidate_count=len(candidate_decisions))
     event_quality = _gate(
-        len(training_events) >= MIN_ACCEPTANCE_EVENTS,
-        scoreable_event_count=len(training_events),
+        len(scoreable_events) >= MIN_ACCEPTANCE_EVENTS,
+        scoreable_event_count=len(scoreable_events),
+        training_event_count=len(training_events),
         minimum=MIN_ACCEPTANCE_EVENTS,
     )
     domain_diversity = _gate(
-        len(domains) >= MIN_ACCEPTANCE_DOMAINS,
-        scoreable_domain_count=len(domains),
+        len(scoreable_domains) >= MIN_ACCEPTANCE_DOMAINS,
+        scoreable_domain_count=len(scoreable_domains),
         minimum=MIN_ACCEPTANCE_DOMAINS,
-        domains=domains,
+        domains=scoreable_domains,
     )
     date_quality = _date_quality(training_events)
     top_tied_count = candidate_decisions[0]["tied_minute_count"] if candidate_decisions else 0