feat(api): add custom voice creation and agent session events (#4013)

## Summary

Adds synchronous and asynchronous voice creation and typed agent session
webhooks. Removes ultrafast from the supported beta agent and session
service tiers.

## Changes

- Add audio.voices.create overloads for an audio sample with consent, or
a text prompt
- Recognize agent.session.created, in_progress, idle, failed, and
action_required webhook events
- Add input_cache_write_12h_tokens to organization usage results
- Remove the ultrafast service_tier literal from beta agent and session
create and update parameters
- Mark the image variations endpoint deprecated and document model
deprecations
- Clarify that voices created from text prompts are supported only in
Live

Co-authored-by: markstuart-oai <323302876+markstuart-oai@users.noreply.github.com>