feat(agents): [1/n] return typed answers from session streams (#4007)

## Summary

Bind a Pydantic output model to the Agents request and its completed
result, so applications can use a typed answer without maintaining
separate schema and parsing code.

Before:
```python
with client.beta.agents.sessions.create(
    agent={"model": MODEL, "text": {"format": {
        "type": "json_schema", "schema": schema,
    }}},
    environment={"type": "none"}, input=QUESTION, stream=True,
) as stream:
    result = stream.get_final_result()
report = Report.model_validate_json(result.output_text)
```

After:
```python
with client.beta.agents.sessions.create(
    agent={"model": MODEL}, environment={"type": "none"},
    input=QUESTION, stream=True, output_type=Report,
) as stream:
    result = stream.get_final_result()
report = result.output_parsed
```

The same `output_type` works on follow-up streams as a local parser for
an already-configured session. `output_parsed` returns the first parsed
final text part; all final text parts are validated. Raw output remains
available; parsing failures expose the completed result through
`AgentOutputParseError.result`. Sync and async helpers share the
existing Pydantic schema normalization and result collection.

Schema conversion follows the existing Responses helper; API errors
report unsupported schema features. Typed tools and outputs can reuse
the same model without output normalization changing its tool argument
schema.

### Stack

- https://github.com/openai/openai-python/pull/4004 (merged
prerequisite)
- https://github.com/openai/openai-python/pull/4007 👈 this PR
- https://github.com/openai/openai-python/pull/4008
- https://github.com/openai/openai-python/pull/4009