feat(task): 添加任务修订功能和超时处理机制

添加了 `revise_task` 路由动作类型，允许用户修改、纠正或重新执行最新活动任务结果。
实现了工具失败指导原则，防止相同类别工具重复失败。
为任务规划器添加了超时处理机制，避免长时间等待。

BREAKING CHANGE: 任务路由逻辑已更新，新增 `revise_task` 动作类型。

fix(api): 修复任务详情API返回完整流程投影

修复了任务详情API端点，现在会包含过滤后的流程运行、事件和工件信息，
并确保时间戳字段正确序列化。

refactor(engine): 优化任务技能解析器摘要节点处理

改进了任务技能解析器对摘要节点的处理逻辑，对于仅依赖文本生成功能的摘要节
点不再分配具体技能，直接使用依赖项输出进行汇总。

test: 增加任务修订和超时处理测试用例

添加了测试用例验证任务修订输入记录反馈、超时回退到单模式以及
摘要节点技能解析等新功能。

app-instance/backend/tests/unit/test_task_skill_resolver.py

@@ -65,8 +65,8 @@ def _publish_skill(workspace: Path, *, skill_name: str) -> None:
     store = SkillSpecStore(workspace)
     draft = DraftService(store).create_new_skill_draft(
         skill_name=skill_name,
-        proposed_content="# API Contract Review\n\nCheck schema compatibility and breaking changes.",
-        proposed_frontmatter={"description": "API contract compatibility review", "tools": []},
+        proposed_content=f"# {skill_name}\n\nCheck schema compatibility and breaking changes.",
+        proposed_frontmatter={"description": f"{skill_name} capability", "tools": []},
         created_by="tester",
         reason="test",
     )
@@ -174,3 +174,51 @@ def test_task_skill_resolver_generates_ephemeral_guidance_when_missing(tmp_path:
     assert reports[0].ephemeral_guidance_id is not None
     assert reports[0].ephemeral_guidance_name == "api-compatibility-review"
     assert reports[0].ephemeral_used is True
+
+
+def test_task_skill_resolver_keeps_summary_nodes_skillless(tmp_path: Path) -> None:
+    _publish_skill(tmp_path, skill_name="multi-search-engine")
+    provider = RecordingProvider(['["multi-search-engine"]'])
+    resolver = TaskSkillResolver(
+        skills_loader=SkillsLoader(tmp_path),
+        draft_service=DraftService(SkillSpecStore(tmp_path)),
+    )
+    graph = ExecutionGraph(
+        strategy="dag",
+        nodes=[
+            ExecutionNode(
+                "summarize",
+                "Compile a clear, concise summary from dependency outputs for the user.",
+                AgentDescriptor(
+                    name="summarize",
+                    metadata={
+                        "skill_query": "Summarization",
+                        "required_capabilities": ["text generation"],
+                    },
+                ),
+                depends_on=["verify_result"],
+                inherited_pinned_skills=["multi-search-engine"],
+                inherited_pinned_skill_contexts=[
+                    SkillContext(name="ephemeral:search-guidance", content="Search again.")
+                ],
+            )
+        ],
+    )
+
+    resolved, reports = asyncio.run(
+        resolver.resolve_graph(
+            graph,
+            task=_task(),
+            user_message="summarize result",
+            attempt_index=2,
+            provider_bundle=_bundle(provider),
+        )
+    )
+
+    assert resolved.nodes[0].inherited_pinned_skills == []
+    assert resolved.nodes[0].inherited_pinned_skill_contexts == []
+    assert resolved.nodes[0].agent.metadata["selected_skill_names"] == []
+    assert reports[0].selected_skill_names == []
+    assert reports[0].ephemeral_used is False
+    assert reports[0].reason == "summary node uses dependency outputs directly"
+    assert provider.calls == []