feat(bot): phase-6 calibration — mock seed, A/B harness, low-trust fallback

- tools/seed_mock.py: idempotent additive mock data (uuid5 ids) covering all
  tables (units, brands, specs, suppliers, materials + EAV/brand/unit links,
  projects, warehouses, inventory, quotes with termsheets/line items/decisions,
  receipts) so phases 6-7 can be exercised end-to-end
- tools/calibrate_matching.py: A/B difflib vs Jev on 35 scenarios (31 positive
  + 4 negative); collects RAW picks with widened bands, classifies against the
  configured thresholds, sweeps EXACT 0.55-0.95 (negatives count as binds),
  recommends FLOOR strictly above the highest wrong pick with margin for
  ±0.05 run-to-run confidence jitter; markdown report + real traces
- matching.py: below-FLOOR picks no longer suppress difflib — they log a
  routing.decision event (source=jev-lowtrust) and leave no result so the
  existing difflib path keeps its chance (0.54 «پردیس» was lost to «new»)
- config.py: JEV_CONF_FLOOR 0.60 -> 0.65 (removes the جعبه→بسته @0.61 wrong
  candidate offer; correct sub-floor picks باق اسپارت/کیلو are difflib-covered)
- test_matching.py: +2 low-trust cases (13/13 mock, 16/16 live)
- docker-compose.yml: mount bot/tools into the bot container

bot/app/config.py

@@ -147,7 +147,9 @@ class Settings(BaseSettings):
     JEV_ENABLED: bool = True
     JEV_TIMEOUT: float = 8.0
     JEV_CONF_EXACT: float = 0.85
-    JEV_CONF_FLOOR: float = 0.60
+    JEV_CONF_FLOOR: float = 0.65   # phase-6 calibration (2026-10-12): 0.60 let one
+                                   # wrong candidate (جعبه→بسته @0.61) through; the
+                                   # correct picks it drops are covered by difflib.
     JEV_MAX_QUESTIONS_PER_FIELD: int = 15
 
     LOG_LEVEL: str = "INFO"

bot/app/matching.py

@@ -168,18 +168,30 @@ async def resolve_mentions(ctx, patch: dict, mentions: list[dict], trace_id) ->
                 key=lambda x: -x["prob"],
             )[:6]
 
+            base = {"field": field, "variants": m["variants"], "confidence": conf,
+                    "choice": (cand.name if cand is not None else None),
+                    "candidates": ranked}
+
             if cand is not None and conf >= settings.JEV_CONF_EXACT:
-                res = {"field": field, "status": "exact", "variants": m["variants"],
-                       "id": cand.id, "name": cand.name, "extra": cand.extra or {},
-                       "confidence": conf}
+                res = {**base, "status": "exact", "id": cand.id, "name": cand.name,
+                       "extra": cand.extra or {}}
             elif cand is not None and conf >= settings.JEV_CONF_FLOOR:
-                res = {"field": field, "status": "ambiguity", "variants": m["variants"],
-                       "id": cand.id, "name": cand.name, "extra": cand.extra or {},
-                       "confidence": conf,
+                res = {**base, "status": "ambiguity", "id": cand.id, "name": cand.name,
+                       "extra": cand.extra or {},
                        "candidates": ranked or [{"id": cand.id, "name": cand.name, "prob": conf}]}
+            elif cand is None:
+                # Explicit «_none» (or an unusable choice): the model judged the whole
+                # list → treat as new (the auto-create path handles it).
+                res = {**base, "status": "new"}
             else:
-                res = {"field": field, "status": "new", "variants": m["variants"],
-                       "confidence": conf, "candidates": ranked}
+                # Picked something but below the floor = low trust: leave NO result so
+                # the difflib fallback still gets its chance (suppressing it here would
+                # let a correct-but-under-confident pick create duplicates instead).
+                await obs.event(trace_id, "routing.decision", field=field, payload={
+                    "source": "jev-lowtrust", "variants": m["variants"],
+                    "choice": cand.name, "confidence": conf,
+                })
+                continue
             for v in m["variants"]:
                 results[normalize_text(v)] = res
             decisions.append({"variants": m["variants"], "status": res["status"],