To ensure we are using correct events, using the realtime types from theopenai sdk. Added conversion tests as well.---[//]: # (BEGIN SAPLING FOOTER)*#1135* __->__#1134

The UI demo is nice, but this let you print more logs to the console (atthe cost of being ugly)

---[//]: # (BEGIN SAPLING FOOTER)*#1141* __->__#1139

Without this, the audio playback contends for CPU with the event queueand mic input, which is bad

MCP recently added `structuredContent` as an additional field in toolresults. This lets us use that field when sending the response to themodel.

#1101)### Problem1. **Incorrect parameter reference**: Suggests passing`output_schema_strict=False` to `Agent()`, but this parameter doesn'texist2. **Vague guidance**: The phrase "make the output type strict" doesn'texplain HOW to make types strict### SolutionUpdated the error message to provide accurate, actionable guidance:**Before:**```"Either make the output type strict, or pass output_schema_strict=False to your Agent()"```**After:**```"Either use a dataclass, Pydantic model, or TypedDict, or wrap your type with AgentOutputSchema(your_type, strict_json_schema=False)"```---------Co-authored-by: Kazuhiro Sera <seratch@openai.com>

## SummaryThis PR implements real-time streaming of function call arguments asrequested in#834. Previously, function call arguments were only emittedafter the entire function call was complete, causing poor userexperience for large parameter generation.## Changes- **Enhanced `ChatCmplStreamHandler`**: Added real-time streaming offunction call arguments during generation- **New streaming logic**: Function call arguments now streamincrementally as they are generated, similar to text content- **Backward compatibility**: Maintains existing behavior for completedfunction calls- **Comprehensive testing**: Added tests for both OpenAI and LiteLLMmodels- **Example implementation**: Created demonstration code showing the newstreaming capabilityCloses#834

### SummaryFixed `MCPUtil.invoke_mcp_tool()` to return `"[]"` instead of logging anerror when MCP tools return empty content. Empty results (`None, []`)are valid outputs and should be treated consistently withfunction tools. The `None -> "[]"` and `[] -> "[]"` behavior comes fromthe MCP Python SDK's[ FastMCP implementation](https://github.com/modelcontextprotocol/python-sdk/blob/6566c08446ad37b66f1457532c7f5a14f243ae10/src/mcp/server/fastmcp/utilities/func_metadata.py#L500).  ### Test plan- Added comprehensive test coverage in`test_mcp_fastmcp_behavior_verification()` that verifies all FastMCPedge cases  - Enhanced FakeMCPServer to support custom content testing scenarios  - Ran full test suite - all tests pass  - Verified the fix handles `None, [], {}, [{}], [[]]` cases correctly  ### Issue numberCloses#1035  ### Checks  - [x] I've added new tests (if relevant)  - [ ] I've added/updated the relevant documentation  - [x] I've run `make lint` and `make format`  - [x] I've made sure tests passCo-authored-by: thein <thein@MacBook-Pro.local>

Breakage from#1142

Add a web demo that uses a fastapi backend. Also remove the CLI UI demo.

Documentation (both written and code)