fix(cma): T4 residual bugs - predict 500 + reports restateDiffClass (cma-t4-residual-bugs-20260916)

B1: /api/cma/dashboard/predict hand-rolled threshold parsing called float() on the whole

expression -> ValueError on multi-clause thresholds like '<=36|>=124' -> 500. Now routes

through app.utils.threshold_parser.eval_threshold (single source of truth; red then yellow,

invalid/blank expr -> False, no crash).

B2: ReportCenter.vue restatement tab template called restateDiffClass while the script

defines restatementDiffClass (name mismatch since ec6af75) -> TypeError on render.

Verified: predict 200 on new code vs 500 on live code; /reports A/B browser run shows 0 vs 3

occurrences of the TypeError; pytest 1113 passed / 1 xfailed; vite build OK.

backend/app/api/dashboard.py

@@ -10,6 +10,7 @@ from app.deps import get_entity_id
 from app.models import KPIDefinition, KPIValue, KPIAlert, User, VoucherDetail
 from app.utils.cache import get as cache_get, set as cache_set
 from app.utils.voucher_guard import apply_voucher_guard
+from app.utils.threshold_parser import eval_threshold
 # 指标方向（反向/正向）唯一权威源 = app.api.kpis.REVERSE_INDICATORS（2026-09-15 M2 准入2 收口）。
 # 本文件原自带一份字面量副本，与 kpis.py 漂移后会与「能力评分」判定相反，故改为导入。
 from app.api.kpis import REVERSE_INDICATORS
@@ -529,17 +530,15 @@ def predict_kpis(db: Session = Depends(get_db), entity_id: int = Depends(get_ent
         predicted_value = a + b * n
         
         # 检查预测值是否触发阈值
+        # 统一走 app.utils.threshold_parser（阈值语法单一事实源，2026-09-16 B1 修复）：
+        # 原实现手工切操作符 + float()，只识别单条件标量表达式；生产库存在多条件
+        # 表达式（如 `<=36|>=124`）/区间/负数 → float() 抛 ValueError → 端点 500。
+        # eval_threshold 对非法/空表达式返回 False（不致命），并支持 A~B、C1|C2、负数。
         alert_level = "none"
-        if k.threshold_red:
-            op = k.threshold_red[:2] if k.threshold_red[1] in "=<>" else k.threshold_red[0]
-            val = float(k.threshold_red.replace(op, "").strip())
-            if (op in (">=", ">") and predicted_value >= val) or (op in ("<=", "<") and predicted_value <= val):
-                alert_level = "red"
-        if alert_level == "none" and k.threshold_yellow:
-            op = k.threshold_yellow[:2] if k.threshold_yellow[1] in "=<>" else k.threshold_yellow[0]
-            val = float(k.threshold_yellow.replace(op, "").strip())
-            if (op in (">=", ">") and predicted_value >= val) or (op in ("<=", "<") and predicted_value <= val):
-                alert_level = "yellow"
+        if eval_threshold(predicted_value, k.threshold_red):
+            alert_level = "red"
+        elif eval_threshold(predicted_value, k.threshold_yellow):
+            alert_level = "yellow"
         
         predictions.append({
             "kpi_id": k.id,