feat(learning): FU-5 — מדידת לולאת-הלמידה מול הכרעות-היו"ר (#133)

מרחיב את halacha_panel_calibrate.py כדי למדוד את הלולאה לאורך-זמן ולשמור
על בריאותה — סוגר את 5 ה-FU של #133.

- --source captured (חדש, אפס-עלות): מצליב סבבי-פאנל שמורים (FU-1) מול
  הכרעות-היו"ר (FU-2) דרך db.panel_rounds_vs_chair, ומדווח split-rate +
  auto-precision + false-keep/false-drop **לכל סבב (per round-day)** מול
  ה-gold-set הגדל. כך רואים את הלולאה עובדת: ככל שהרובריקה משתפרת
  (FU-4 → אימוץ-יו"ר) — precision נשמר ו-split יורד. בלי re-vote, בלי LLM.
- summarize_calibration() + bucket_by_round() — עוזרים טהורים (offline-
  testable). משתפים את analyze_pairs של FU-4 → "מה נכשל" מחושב במקום אחד
  (בלי drift, G2).
- anon-stability: שתי המדידות מדווחות את שיעור-יציבות מבחן-האנונימיזציה
  (#81.7) כמטריקת-בריאות נגד echo-chamber — נפילה = שינון במקום היגיון.
- --source live (קיים): נוסף עמוד split-rate מפורש + anon-stability.
- tests/test_panel_calibrate_captured.py — 5 בדיקות offline. SCRIPTS.md
  עודכן. smoke read-only עבר (0 זוגות → nothing-to-measure).

Invariants: read-only מדידה · INV-G10 (האמת=הכרעת-יו"ר) · anti-echo-
chamber (anon-stability) · G2 (analyze_pairs מקור-יחיד). רגרסיה 30 עברו.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

mcp-server/tests/test_panel_calibrate_captured.py

@@ -0,0 +1,75 @@
+"""Tests for #133 / FU-5 — captured-mode calibration of the halacha panel.
+
+Covers the PURE helpers in scripts/halacha_panel_calibrate.py
+(summarize_calibration, bucket_by_round): from captured (panel ⋈ chair) pairs
+they must report the split-rate and auto-precision the panel ACTUALLY delivered
+against the chair's ground-truth, and break it down per round-day so the loop's
+trend is visible. Fully OFFLINE (no DB, no LLM, no re-voting).
+"""
+
+from __future__ import annotations
+
+import sys
+from pathlib import Path
+
+sys.path.insert(0, str(Path(__file__).resolve().parents[2] / "scripts"))
+import halacha_panel_calibrate as cal  # noqa: E402
+
+
+def _pair(chair, verdict, action, ts="2026-06-12T04:00:00Z"):
+    # judge votes are irrelevant to split-rate/precision here; keep them aligned
+    return {
+        "chair_keep": chair, "verdict": verdict, "applied_action": action,
+        "round_ts": ts, "rule_statement": "r",
+        "claude_vote": chair, "claude_reason": "",
+        "deepseek_vote": chair, "deepseek_reason": "",
+        "gemini_vote": chair, "gemini_reason": "",
+    }
+
+
+def test_split_rate_and_precision():
+    pairs = [
+        _pair(True, "unanimous_yes", "approved"),   # auto-correct
+        _pair(False, "unanimous_no", "rejected"),   # auto-correct
+        _pair(False, "unanimous_yes", "approved"),  # auto WRONG (false-keep)
+        _pair(True, "split", "chair"),              # escalated
+    ]
+    s = cal.summarize_calibration(pairs)
+    assert s["n"] == 4
+    assert s["escalated"] == 1
+    assert s["auto_decided"] == 3
+    assert s["split_rate"] == 0.25
+    # 3 auto-decisions, 1 wrong (false-keep) → precision 2/3
+    assert s["false_keep"] == 1 and s["false_drop"] == 0
+    assert round(s["auto_precision"], 2) == 0.67
+
+
+def test_empty_pairs_safe():
+    s = cal.summarize_calibration([])
+    assert s["n"] == 0
+    assert s["split_rate"] is None and s["auto_precision"] is None
+
+
+def test_unlabeled_pairs_filtered():
+    s = cal.summarize_calibration([_pair(None, "split", "chair")])
+    assert s["n"] == 0  # chair=None contributes no calibration signal
+
+
+def test_bucket_by_round_trend():
+    pairs = [
+        _pair(True, "unanimous_yes", "approved", ts="2026-06-10T04:00:00Z"),
+        _pair(False, "split", "chair", ts="2026-06-10T04:00:00Z"),
+        _pair(True, "unanimous_yes", "approved", ts="2026-06-12T05:00:00Z"),
+    ]
+    trend = cal.bucket_by_round(pairs)
+    days = [d for d, _ in trend]
+    assert days == ["2026-06-10", "2026-06-12"]  # sorted by day
+    assert trend[0][1]["n"] == 2 and trend[0][1]["split_rate"] == 0.5
+    assert trend[1][1]["n"] == 1 and trend[1][1]["split_rate"] == 0.0
+
+
+def test_missing_round_ts_bucketed_unknown():
+    p = _pair(True, "unanimous_yes", "approved")
+    del p["round_ts"]
+    trend = cal.bucket_by_round([p])
+    assert trend[0][0] == "unknown"