tts : add support for SparkTTS

[ecyht2]

Prerequisites

• I am running the latest code. Mention the version if possible as well.
• I carefully followed the README.md.
• I searched using keywords relevant to my issue to make sure that I am creating a new issue that is not already open (or closed).
• I reviewed the Discussions, and have a new and useful enhancement to share.

Feature Description

HF: https://huggingface.co/SparkAudio/Spark-TTS-0.5B

Motivation

A new TTS model that can be supported by llama.cpp.

Possible Implementation

I might be wrong here but it seems like SparkTTS has a simlar architecture as OuteTTS and Orpheus TTS (#12476) but it uses Qwen2.5-0.5B.

They are using their own audio decoder called BiCodec. Sample python implementation: https://github.com/SparkAudio/Spark-TTS/blob/main/sparktts/models/bicodec.py

Similar model support (OuteTTS): #10784
Can be used as a reference how to implement this.