test(cma): clear 10 pre-existing baseline failures (cma-baseline-tests-cleanup-20260915)

No production code changed. Root causes were all test-side staleness:

- test_cash TestVoucherImport: fixture filename contained 'test' -> rejected by the intentional import guard (d71c7fe); renamed, and missing-columns case now asserts the column check is what returns 400

- test_budget_tech_improve TestP1ValueCollect: seeded voucher_details source rows; collector deliberately skips empty aggregates since d71c7fe (no more implicit 0), and preview call now passes an explicit period

- test_roadmap_r2r5 / test_budget / test_ai_suggestions: seeds hardcoded 2026-08 (authoring month) while detectors/rule engine use the current month -> replaced with current-period seeds

pytest tests/ -q -rf: 1094 passed, 1 xfailed, 0 failed (baseline was 10 failed, 1084 passed, 1 xfailed)

backend/tests/test_ai_suggestions.py

@@ -231,9 +231,12 @@ class TestRuleSuggestions:
         create_test_user(db)
         token = get_token_for_user(client)
         kpi = create_test_kpi(db, target_value=100.0)
-        db.add(KPIValue(kpi_id=kpi.id, period="2026-08", actual_value=200.0))
-        db.add(BudgetPlan(entity_id=1, kpi_id=kpi.id, period="2026-08", budget_value=100.0,
-                          budget_year=2026, budget_month=8, status="active"))
+        # 规则引擎按「当月」取预算（ai_analysis.py: _generate_rule_suggestions 的 period）
+        # 原用例写死 2026-08（编写时的当月），跨月后与实现错配 → 恒 0 条（2026-09 跨月实证）
+        period = datetime.now().strftime("%Y-%m")
+        db.add(KPIValue(kpi_id=kpi.id, period=period, actual_value=200.0))
+        db.add(BudgetPlan(entity_id=1, kpi_id=kpi.id, period=period, budget_value=100.0,
+                          budget_year=int(period[:4]), budget_month=int(period[5:7]), status="active"))
         db.commit()
 
         resp = client.get("/api/cma/ai/dashboard-analysis", headers=auth_header(token))