fix: predictions row carries arm run_id under shared gate_log; drain evolve gather on failure (algo #6)

tests/unit/test_inference_pair_aggregate.py

@@ -332,8 +332,8 @@ class TestRunInferencePairEndToEnd:
         assert result.total == 1
         assert result.correct == 1
 
-        # 逐题溯源：predictions 表两条 record 都在
-        rows = harness_log.query("SELECT * FROM predictions WHERE run_id = ?", ("test-run",))
+        # 逐题溯源：predictions 表两条 record 都在（record 显式携带传入的 run_id）
+        rows = harness_log.query("SELECT * FROM predictions WHERE run_id = ?", ("run-pair-e2e",))
         qids = {r["question_id"] for r in rows}
         assert qids == {"po", "pm"}
 
@@ -360,6 +360,6 @@ class TestRunInferencePairEndToEnd:
         )
 
         assert result.total == 1  # single 存活，孤儿剔除
-        # 逐题溯源：孤儿题仍逐题落库（推理不变）
-        rows = harness_log.query("SELECT * FROM predictions WHERE run_id = ?", ("test-run",))
+        # 逐题溯源：孤儿题仍逐题落库（推理不变；record 显式携带传入的 run_id）
+        rows = harness_log.query("SELECT * FROM predictions WHERE run_id = ?", ("run-orphan-e2e",))
         assert {r["question_id"] for r in rows} == {"s1", "po"}