test: remove LangChain invoke methods integration test and rename test toolsets

Removed the comprehensive LangChain invoke methods integration test file which was testing various invocation patterns (astream_events, astream, stream, invoke, ainvoke) with both streaming and non-streaming scenarios. Also renamed TestToolSet to SampleToolSet across multiple integration test files to avoid confusion with actual test classes.

The removed test file contained extensive setup for mocking OpenAI protocol services and comprehensive test cases for different LangChain/LangGraph invocation methods including streaming protocols and tool call scenarios.

The toolset renaming affects AgentScope, CrewAI, Google ADK, LangChain, LangGraph, and PydanticAI integration tests to maintain consistency and prevent naming conflicts.

feat(browser-toolset): add thread awareness and error handling for Playwright connections

Added comprehensive thread awareness to BrowserToolSet's Playwright connection management. The implementation now tracks which thread created the Playwright connection and properly recreates it when the creating thread exits, preventing greenlet binding errors in multi-threaded environments like LangGraph ToolNodes.

Also added proper error handling for 'cannot switch to a different thread' exceptions, automatically resetting both Playwright connections and sandbox instances when such errors occur to prevent cascading failures.

The changes include new test coverage for concurrent access, thread lifecycle management, and error recovery scenarios to ensure robust operation in production environments.

test(browser-toolset): enhance error handling tests with thread awareness

Expanded the browser toolset error handling tests to cover thread-aware Playwright connection management. Added comprehensive test cases for concurrent access patterns, thread lifecycle scenarios, and specific error handling for greenlet binding issues that occur when switching between threads in LangGraph environments.

The new tests verify proper connection reuse within the same thread, correct recreation when the creating thread exits, and appropriate error handling for cross-thread Playwright operations.

chore(test): remove duplicate LangChain invoke methods test file

Removed the duplicate test file that appears to be an older version of the LangChain invoke methods tests, maintaining only the comprehensive version that covers all invocation patterns and streaming scenarios.

test: 移除 LangChain 调用方法集成测试并重命名测试工具集

移除了全面的 LangChain 调用方法集成测试文件，该文件测试了各种调用模式（astream_events、astream、stream、invoke、ainvoke）以及流式和非流式场景。同时在多个集成测试文件中将 TestToolSet 重命名为 SampleToolSet，以避免与实际测试类混淆。

被删除的测试文件包含大量用于模拟 OpenAI 协议服务的设置，以及针对不同 LangChain/LangGraph 调用方法的全面测试用例，包括流式协议和工具调用场景。

工具集重命名影响了 AgentScope、CrewAI、Google ADK、LangChain、LangGraph 和 PydanticAI 集成测试，以保持一致性并防止命名冲突。

feat(browser-toolset): 为 Playwright 连接添加线程感知和错误处理功能

为 BrowserToolSet 的 Playwright 连接管理添加了全面的线程感知功能。实现现在跟踪哪个线程创建了 Playwright 连接，并在线程退出时正确重新创建它，防止在多线程环境（如 LangGraph ToolNodes）中的 greenlet 绑定错误。

还为 'cannot switch to a different thread' 异常添加了适当的错误处理，在发生此类错误时自动重置 Playwright 连接和沙箱实例，以防止级联故障。

更改包括对并发访问、线程生命周期管理和错误恢复场景的新测试覆盖，以确保在生产环境中稳健运行。

test(browser-toolset): 使用线程感知增强错误处理测试

扩展浏览器工具集错误处理测试以涵盖线程感知的 Playwright 连接管理。添加了对并发访问模式、线程生命周期场景的全面测试用例，以及针对在 LangGraph 环境中跨线程切换时发生的 greenlet 绑定问题的具体错误处理。

新测试验证同一线程内的正确连接复用、创建线程退出时的正确重建，以及跨线程 Playwright 操作的适当错误处理。

chore(test): 删除重复的 LangChain 调用方法测试文件

删除似乎是 LangChain 调用方法测试旧版本的重复测试文件，只保留涵盖所有调用模式和流式场景的完整版本。

Change-Id: I4d963b9b041bf81cc2f5f20d3c2f7cd47abe6bbb
Signed-off-by: OhYee <oyohyee@oyohyee.com>

Author: OhYee

…n/langchain/test_agent_invoke_methods.py …n/test_agent_invoke_methods_unittests.pytests/unittests/integration/langchain/test_agent_invoke_methods.py renamed to tests/unittests/integration/langchain/test_agent_invoke_methods_unittests.py tests/unittests/integration/langchain/test_agent_invoke_methods.py renamed to tests/unittests/integration/langchain/test_agent_invoke_methods_unittests.py

@@ -21,7 +21,7 @@
 from .scenarios import Scenarios
 
 
-class TestToolSet(CommonToolSet):
+class SampleToolSet(CommonToolSet):
     """测试用工具集"""
 
     def __init__(self, timezone: str = "UTC"):
@@ -150,9 +150,9 @@ def mocked_model(
         return model("mock-model")
 
     @pytest.fixture
-    def mocked_toolset(self) -> TestToolSet:
+    def mocked_toolset(self) -> SampleToolSet:
         """创建 mock 的工具集"""
-        return TestToolSet(timezone="UTC")
+        return SampleToolSet(timezone="UTC")
 
     # =========================================================================
     # 测试：简单对话（无工具调用）
@@ -194,7 +194,7 @@ async def test_multi_tool_calls(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试多工具同时调用"""
         # 使用默认的多工具场景
@@ -223,7 +223,7 @@ async def test_stream_options_validation(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试 stream_options 在请求中的正确性"""
         # 使用默认场景

tests/unittests/integration/test_agentscope.py

@@ -337,7 +337,9 @@ def toolset(self, mock_sandbox):
             ts.lock = threading.Lock()
             return ts
 
-    def test_get_playwright_records_creating_thread(self, toolset, mock_sandbox):
+    def test_get_playwright_records_creating_thread(
+        self, toolset, mock_sandbox
+    ):
         """测试 _get_playwright 记录创建连接的线程"""
         toolset._get_playwright(mock_sandbox)
 

tests/unittests/integration/test_browser_toolset_error_handling.py

@@ -23,7 +23,7 @@
 from .scenarios import Scenarios
 
 
-class TestToolSet(CommonToolSet):
+class SampleToolSet(CommonToolSet):
     """测试用工具集"""
 
     def __init__(self, timezone: str = "UTC"):
@@ -149,9 +149,9 @@ def mocked_model(
         return model("mock-model")
 
     @pytest.fixture
-    def mocked_toolset(self) -> TestToolSet:
+    def mocked_toolset(self) -> SampleToolSet:
         """创建 mock 的工具集"""
-        return TestToolSet(timezone="UTC")
+        return SampleToolSet(timezone="UTC")
 
     # =========================================================================
     # 测试：简单对话（无工具调用）
@@ -191,7 +191,7 @@ def test_multi_tool_calls(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试多工具同时调用
 
@@ -210,7 +210,7 @@ def test_stream_options_validation(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试 stream_options 在请求中的正确性
 

tests/unittests/integration/test_crewai.py

@@ -21,7 +21,7 @@
 from .scenarios import Scenarios
 
 
-class TestToolSet(CommonToolSet):
+class SampleToolSet(CommonToolSet):
     """测试用工具集"""
 
     def __init__(self, timezone: str = "UTC"):
@@ -208,9 +208,9 @@ def mocked_model(
         return model("mock-model")
 
     @pytest.fixture
-    def mocked_toolset(self) -> TestToolSet:
+    def mocked_toolset(self) -> SampleToolSet:
         """创建 mock 的工具集"""
-        return TestToolSet(timezone="UTC")
+        return SampleToolSet(timezone="UTC")
 
     # =========================================================================
     # 测试：简单对话（无工具调用）
@@ -252,7 +252,7 @@ async def test_single_tool_call(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试单次工具调用"""
         # 配置场景
@@ -284,7 +284,7 @@ async def test_multi_tool_calls(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试多工具同时调用"""
         # 使用默认的多工具场景
@@ -315,7 +315,7 @@ async def test_stream_options_validation(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试 stream_options 在请求中的正确性"""
         # 使用默认场景

tests/unittests/integration/test_google_adk.py

@@ -22,7 +22,7 @@
 from .scenarios import Scenarios
 
 
-class TestToolSet(CommonToolSet):
+class SampleToolSet(CommonToolSet):
     """测试用工具集"""
 
     def __init__(self, timezone: str = "UTC"):
@@ -201,9 +201,9 @@ def mocked_model(
         return model("mock-model")
 
     @pytest.fixture
-    def mocked_toolset(self) -> TestToolSet:
+    def mocked_toolset(self) -> SampleToolSet:
         """创建 mock 的工具集"""
-        return TestToolSet(timezone="UTC")
+        return SampleToolSet(timezone="UTC")
 
     # =========================================================================
     # 测试：简单对话（无工具调用）
@@ -244,7 +244,7 @@ def test_single_tool_call(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试单次工具调用"""
         # 配置场景
@@ -276,7 +276,7 @@ def test_multi_tool_calls(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试多工具同时调用"""
         # 使用默认的多工具场景
@@ -307,7 +307,7 @@ def test_stream_options_in_requests(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试请求中的 stream_options 设置"""
         from langchain_openai import ChatOpenAI
@@ -324,7 +324,7 @@ def test_stream_options_validation(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试 stream_options 在请求中的正确性"""
         # 使用默认场景
@@ -370,7 +370,7 @@ async def test_async_invoke(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试异步调用"""
         # 使用默认场景

tests/unittests/integration/test_langchain.py

@@ -22,7 +22,7 @@
 from .scenarios import Scenarios
 
 
-class TestToolSet(CommonToolSet):
+class SampleToolSet(CommonToolSet):
     """测试用工具集"""
 
     def __init__(self, timezone: str = "UTC"):
@@ -233,9 +233,9 @@ def mocked_model(
         return model("mock-model")
 
     @pytest.fixture
-    def mocked_toolset(self) -> TestToolSet:
+    def mocked_toolset(self) -> SampleToolSet:
         """创建 mock 的工具集"""
-        return TestToolSet(timezone="UTC")
+        return SampleToolSet(timezone="UTC")
 
     # =========================================================================
     # 测试：简单对话（无工具调用）
@@ -275,7 +275,7 @@ def test_single_tool_call(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试单次工具调用"""
         # 配置场景
@@ -306,7 +306,7 @@ def test_multi_tool_calls(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试多工具同时调用"""
         # 使用默认的多工具场景
@@ -336,7 +336,7 @@ def test_stream_options_validation(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试 stream_options 在请求中的正确性"""
         # 使用默认场景
@@ -377,7 +377,7 @@ async def test_async_invoke(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试异步调用"""
         # 使用默认场景

tests/unittests/integration/test_langgraph.py

@@ -22,7 +22,7 @@
 from .scenarios import Scenarios
 
 
-class TestToolSet(CommonToolSet):
+class SampleToolSet(CommonToolSet):
     """测试用工具集"""
 
     def __init__(self, timezone: str = "UTC"):
@@ -203,9 +203,9 @@ def mocked_model(
         return model("mock-model")
 
     @pytest.fixture
-    def mocked_toolset(self) -> TestToolSet:
+    def mocked_toolset(self) -> SampleToolSet:
         """创建 mock 的工具集"""
-        return TestToolSet(timezone="UTC")
+        return SampleToolSet(timezone="UTC")
 
     # =========================================================================
     # 测试：简单对话（无工具调用）
@@ -245,7 +245,7 @@ def test_single_tool_call(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试单次工具调用"""
         # 配置场景
@@ -276,7 +276,7 @@ def test_multi_tool_calls(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试多工具同时调用"""
         # 使用默认的多工具场景
@@ -305,7 +305,7 @@ def test_stream_options_validation(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试 stream_options 在请求中的正确性
 
@@ -349,7 +349,7 @@ async def test_async_invoke(
         self,
         mock_server: MockLLMServer,
         mocked_model: CommonModel,
-        mocked_toolset: TestToolSet,
+        mocked_toolset: SampleToolSet,
     ):
         """测试异步调用"""
         # 使用默认场景