### Check for existing issues

- [x] I have searched the existing issues and checked that my issue is not a duplicate.

### What happened?

Errors seems not to be re-raised mid stream while using the Responses API with openai models. I observe the behaviour with : - Missing API key - Context windows exceeded

Instead of re-raising the error, the error or just yield as event :

```python type=<ResponsesAPIStreamEvents.RESPONSE_FAILED: 'response.failed'> response=ResponsesAPIResponse(id='resp_bGl0ZWxsbTpjdXN0b21fbGxtX3Byb3ZpZGVyOm9wZW5haTttb2RlbF9pZDpOb25lO3Jlc3BvbnNlX2lkOnJlc3BfMDNlYmFhMzk3N2I0NmY2ZjAwNmExODhjOGIyZDQ4ODE5ZmI2NzVjN2QzMzE3ZTIyNWE=', created_at=1779993739, error={'code': 'context_length_exceeded', 'message': 'Your input exceeds the context window of this model. Please adjust your input and try again.'}, incomplete_details=None, instructions=None, metadata={}, model='gpt-5.4-mini-2026-03-17', object='response', output=[], parallel_tool_calls=True, temperature=1.0, tool_choice='auto', tools=[], top_p=0.98, max_output_tokens=None, previous_response_id=None, reasoning={'context': 'current_turn', 'effort': 'none', 'summary': None}, status='failed', text={'format': {'type…