fix(fetch-caliber): 取数层排除半年/预测伪期间 + subjects 响应补 entity_id

R3 期间口径（仅 YYYY-MM）延伸到取数层：新增 alert_entity_scope.month_period_clause()（跨库可移植 SQL 等价式），应用到 15 个文件 26 处「按 period 倒序取最新值/最近N期」位点；dashboard 原 MySQL-only REGEXP 正则收口到同单一事实源。

GET /api/cma/subjects 响应增补 entity_id（NULL=全局标准准则科目 / N=账套私有明细），既有 10 字段语义与名称不变。

证据：pending-tasks/_reports/yanxue-20260924-cma-fetch-caliber-fixes-fullstack.md（改前/改后 A/B：F_NET_PROFIT 2026-H2/-55.0 → 2026-08/-10.46）。新增 7 例回归测试；全量 1 failed(预存在) 1473 passed，基线同项。

backend/app/api/predict.py

@@ -11,6 +11,7 @@ from app.utils.cash_forecast_engine import (
 )
 from app.database import get_db
 from app.deps import get_entity_id, resolve_entity_for_request, assert_entity_scope
+from app.utils.alert_entity_scope import month_period_clause
 from app.auth_middleware import require_auth_401
 from sqlalchemy.orm import Session
 
@@ -772,7 +773,7 @@ def _growth_quality_entity_from_db(db: Session, entity_id: int, period=None):
     )
     if period:
         q = q.filter(KPIValue.period == period)
-    rows = q.order_by(KPIValue.period.desc()).all()
+    rows = q.filter(month_period_clause(KPIValue.period)).order_by(KPIValue.period.desc()).all()
 
     entity = {"entity": ent.name}
     seen = set()