feat(agent): add redis short-term user context cache and align tests

docs/bugs/2026-03-05-agent-runtime-bugs.md

@@ -40,7 +40,7 @@ backend/src/core/agent/infrastructure/litellm/usage_tracker.py:26
 
 ### 复现步骤
 1. 重启服务: `infra/scripts/app.sh stop && infra/scripts/app.sh start`
-2. 运行诊断: `uv run python test_agent_sse_flow.py`
+2. 运行诊断: `AGENT_LIVE_INTEGRATION=1 uv run pytest backend/tests/integration/v1/agent/test_sse_flow_live.py -m live -v`
 
 ### 影响范围
 - LLM 调用成功，但无法提取 token 使用量和成本
@@ -94,7 +94,7 @@ register_model({
 ### 验证方法
 修复后运行:
 ```bash
-uv run python test_agent_sse_flow.py
+AGENT_LIVE_INTEGRATION=1 uv run pytest backend/tests/integration/v1/agent/test_sse_flow_live.py -m live -v
 ```
 预期: 
 - 看到 `RUN_STARTED` 和 `RUN_FINISHED` 事件
@@ -112,7 +112,7 @@ uv run python test_agent_sse_flow.py
 ~~**HIGH** - 阻塞 CI/CD 流程~~ **已解决**
 
 ### 问题描述
-`test_agent_closed_loop_live.py` 测试在 120 秒后超时，未完成执行。
+`test_sse_flow_live.py` 测试在 120 秒后超时，未完成执行。
 
 ### 根本原因
 - **阶段 1**: 由 Bug #1 引起（LLM Provider 配置错误）- **已修复**
@@ -128,7 +128,7 @@ Bug #1 和 #1.1 修复后，测试应能正常完成。
 ### 复现步骤
 ```bash
 cd .worktrees/feature-agent-runtime-closed-loop
-AGENT_LIVE_E2E=1 uv run pytest backend/tests/e2e/test_agent_closed_loop_live.py -m live -v
+AGENT_LIVE_INTEGRATION=1 uv run pytest backend/tests/integration/v1/agent/test_sse_flow_live.py -m live -v
 ```
 
 ### 预期行为
@@ -187,7 +187,7 @@ AGENT_LIVE_E2E=1 uv run pytest backend/tests/e2e/test_agent_closed_loop_live.py
    - [ ] 重启服务验证
 
 3. [ ] **验证修复**
-   - [ ] 运行 `test_agent_sse_flow.py`
+   - [ ] 运行 `test_sse_flow_live.py`
    - [ ] 确认事件流完整（RUN_STARTED → RUN_FINISHED）
    - [ ] 检查 DB 留痕
 
@@ -215,10 +215,10 @@ infra/scripts/app.sh start
 curl http://localhost:5775/health  # 成功
 
 # 3. 运行 live E2E (超时)
-AGENT_LIVE_E2E=1 uv run pytest backend/tests/e2e/test_agent_closed_loop_live.py -m live -v
-AGENT_lIVE_e2e=1 uv run pytest backend/tests/e2e/test_agent_closed_loop_live.py -m live -v
+AGENT_LIVE_INTEGRATION=1 uv run pytest backend/tests/integration/v1/agent/test_sse_flow_live.py -m live -v
+AGENT_LIVE_INTEGRATION=1 uv run pytest backend/tests/integration/v1/agent/test_sse_flow_live.py -m live -v
 # 超时
-uv run python test_agent_sse_flow.py  # 失败 (LLM Provider 错误)
+AGENT_LIVE_INTEGRATION=1 uv run pytest backend/tests/integration/v1/agent/test_sse_flow_live.py -m live -v  # 失败 (LLM Provider 错误)
 
 # 5. 检查日志
 tail -f logs/worker-default.log  # 发现根本原因
@@ -233,7 +233,7 @@ infra/scripts/app.sh stop && infra/scripts/app.sh start
 curl http://localhost:5775/health  # 成功
 
 # 9. 运行诊断脚本
-uv run python test_agent_sse_flow.py  # 失败 (模型定价未映射)
+AGENT_LIVE_INTEGRATION=1 uv run pytest backend/tests/integration/v1/agent/test_sse_flow_live.py -m live -v  # 失败 (模型定价未映射)
 
 # 10. 检查日志
 tail -f logs/worker-default.log  # 发现新错误: 模型未映射
@@ -285,7 +285,7 @@ tail -f logs/worker-default.log  # 发现新错误: 模型未映射
 
 **命令**:
 ```bash
-uv run python test_agent_sse_flow.py
+AGENT_LIVE_INTEGRATION=1 uv run pytest backend/tests/integration/v1/agent/test_sse_flow_live.py -m live -v
 ```
 
 **结果**: ✅ **成功**
@@ -363,6 +363,6 @@ uv run python test_agent_sse_flow.py
 ### 测试覆盖
 修复后需重新运行完整测试套件:
 ```bash
-uv run python test_agent_sse_flow.py
-AGENT_LIVE_E2E=1 uv run pytest backend/tests/e2e/test_agent_closed_loop_live.py -m live -v
+AGENT_LIVE_INTEGRATION=1 uv run pytest backend/tests/integration/v1/agent/test_sse_flow_live.py -m live -v
+AGENT_LIVE_INTEGRATION=1 uv run pytest backend/tests/integration/v1/agent/test_sse_flow_live.py -m live -v
 ```

docs/todo/todo.md

@@ -1,2 +0,0 @@
-1. memory短期的加载。memory的生命周期为ttl+对话条目+session_id。用crewai
-2.