fix(intelligent-eval): trigger worker/planner with isolated session

t480 排查：评估 cd636d71 一直'等待 OpenClaw 创建会话'，scan loop 每 60s 触发
worker，但 worker 被触发后 0 次工具调用、直接幻觉输出'评估 pending_approval'
（实际 executing），任务永不认领。

根因：openclaw agent --agent main 复用 main 持久 session，多次触发累积上下文
缓存（~12 万 token）后 LLM 不再执行 worker skill 的 API 步骤。
验证：独立 --session-id 触发 worker → 正常取任务、建会话、close。

修复：worker/planner 触发命令加 --session-id（每次唯一 agenteval-worker-*/planner-*），
避免 main session 污染；timeout 300→600（独立 session 首次加载 skill 更慢）。
测试：触发断言含 --session-id，901 passed

backend/agenteval/web/app.py

@@ -2,6 +2,7 @@
 
 import asyncio
 import logging
+import time
 from contextlib import asynccontextmanager
 from pathlib import Path
 
@@ -213,13 +214,18 @@ async def _trigger_intelligent_worker() -> bool:
                 "agent",
                 "--agent",
                 "main",
+                # 必须用独立 session：`--agent main` 复用 main 持久 session，多次触发
+                # 累积上下文缓存（~12 万 token）后 LLM 不再执行 worker skill 的 API
+                # 步骤，直接幻觉输出（如"评估 pending_approval"）而不取任务建会话。
+                "--session-id",
+                f"agenteval-worker-{int(time.time())}",
                 "-m",
                 worker_msg,
                 "--json",
             ],
             capture_output=True,
             text=True,
-            timeout=300,
+            timeout=600,
         )
         _logger.info("Worker 触发完成 exit=%s", proc.returncode)
         if proc.returncode != 0:
@@ -264,13 +270,16 @@ async def _trigger_intelligent_planner() -> bool:
                 "agent",
                 "--agent",
                 "main",
+                # 同 worker：独立 session 避免 main 持久 session 上下文缓存污染
+                "--session-id",
+                f"agenteval-planner-{int(time.time())}",
                 "-m",
                 planner_msg,
                 "--json",
             ],
             capture_output=True,
             text=True,
-            timeout=300,
+            timeout=600,
         )
         _logger.info("Planner 触发完成 exit=%s", proc.returncode)
         if proc.returncode != 0:

tests/integration/test_intelligent_eval_scan_scheduler.py

@@ -79,6 +79,9 @@ def test_trigger_worker_calls_docker_exec(monkeypatch):
     assert "docker" in joined and "openclaw" in joined
     assert "agenteval-intelligent-worker" in joined
     assert "--agent" in joined and "main" in joined
+    # 独立 session：避免 main 持久 session 上下文缓存污染导致 worker 幻觉不执行
+    assert "--session-id" in joined
+    assert "agenteval-worker-" in joined
 
 
 def test_trigger_worker_msg_has_execute_semantics(monkeypatch):
@@ -154,6 +157,9 @@ def test_trigger_planner_calls_docker_exec(monkeypatch):
     joined = " ".join(calls[0])
     assert "docker" in joined and "openclaw" in joined
     assert "agenteval-intelligent-planner" in joined
+    # 独立 session（同 worker 防上下文缓存污染）
+    assert "--session-id" in joined
+    assert "agenteval-planner-" in joined
     assert "立即完成当前任务" in joined
     assert "不要等待下一节拍" in joined