Support Realtime custom voice objects

[lionel-oai]

Summary

This PR fixes Realtime custom voice handling in the Agents SDK.

Realtime sessions can receive and send structured custom voice objects such as {"id": "voice_..."}, but the SDK previously typed voice settings as strings and validated inbound server events before updating response lifecycle state. If a server event such as response.created or response.done contained a structured voice object that failed validation, the SDK could skip response state updates and leave the response-create sequencer blocked. That could prevent the next response.create from being sent after tool output.

The change adds typed support for custom voice objects in Realtime session settings, preserves structured voices when building outbound session.update payloads, and adds a validation fallback for inbound server events so custom voice objects do not break response lifecycle tracking.

Tests

• make format
• make lint
• uv run pytest -q tests/realtime/test_openai_realtime.py tests/realtime/test_realtime_model_settings.py
• uv run pytest -q tests/realtime/test_session.py -k "handoff_session_update_preserves_custom_voice or handoff_tool_handling"
• uv run mypy src/agents/realtime/config.py src/agents/realtime/openai_realtime.py tests/realtime/test_openai_realtime.py
• uv run pyright src/agents/realtime/config.py src/agents/realtime/openai_realtime.py tests/realtime/test_openai_realtime.py
• uv run mypy tests/realtime/test_session.py
• uv run pyright tests/realtime/test_session.py

Full make tests / make typecheck were not completed locally because optional dependency installation was blocked by a socket-firewall tunnel failure while downloading docstring-parser==0.18.0.

[seratch]

If we upgrade openai package to openai>=2.36.0 , this workaround is not necessary while _normalize_custom_voice_for_server_event_validation is still required even with the latest version.

Can you add quick TODO comments explaining why and when to remove to these internal workarounds?