Add Soniox TTS plugin

[matejmarinko-soniox]

Summary

Adds Text-to-Speech to the existing livekit-plugins-soniox package via Soniox's real-time TTS WebSocket API.

• Streaming synthesis with progressive text input (LLM chunks forwarded as they arrive)
• One persistent WebSocket shared across streams.
• Configurable model, language, voice, audio_format, sample_rate, bitrate; update_options() for runtime changes.

Usage

from livekit.plugins import soniox

session = AgentSession(
    tts=soniox.TTS(language="en", voice="Maya"),
    # ...
)

Reads SONIOX_API_KEY from env, or pass api_key= explicitly.

[devin-ai-integration]

✅ Devin Review: No Issues Found

Devin Review analyzed this PR and found no bugs or issues to report.

[aasishkc44]

Hey, When is this expected to be merged?

[tinalenguyen]

i left a small comment, otherwise lgtm and works well! neat implementation 😸