Implement complete Kokoro TTS integration with proper API and voice support

[Copilot]

This PR implements a complete integration of Kokoro TTS into the ebook2audiobook project, following the guidance from the official Kokoro GitHub repository (https://github.com/hexgrad/kokoro).

Key Changes

Proper API Implementation:

• Integrated the real Kokoro TTS package using KModel and KPipeline classes
• Replaced placeholder code with actual Kokoro API calls
• Added proper error handling and dependency management

Voice Support:

• Added 5 built-in Kokoro voices: af_heart, af_bella, af_sarah, am_adam, am_michael
• Implemented voice selection functionality through the existing voice management system
• Added voice cloning support with fallback to built-in Kokoro voices

Engine Integration:

• Added KOKORO engine to TTS_ENGINES configuration
• Updated tts_manager.py to route Kokoro requests to the new implementation
• Configured proper model settings and sample rates (24kHz)

Dependencies:

• Installed Kokoro TTS package directly from GitHub
• Added all required dependencies: torch, torchaudio, numpy, soundfile, huggingface_hub

Testing Results

✅ Basic synthesis: Successfully generates audio from text
✅ Voice selection: All 5 voices work correctly with different audio characteristics
✅ TTS Manager integration: Seamlessly works through the existing TTS management system
✅ File output: Generates proper audio files in FLAC format
✅ Voice cloning: Implements cloning with graceful fallback when Coqui TTS unavailable

The integration maintains compatibility with the existing ebook2audiobook architecture while providing access to Kokoro's high-quality neural TTS capabilities.

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

[DrewThomasson]

@copilot
Here is the github repo for kokoro tts if it helps you

https://github.com/hexgrad/kokoro

[Copilot]

@copilot
Here is the github repo for kokoro tts if it helps you

...

Thank you for the GitHub repository link! I've now installed the actual Kokoro TTS package and will update the implementation to use the correct API (KModel and KPipeline classes) instead of the placeholder code I had before.