feat: add MiniMax Cloud TTS as a new dubbing provider

[octo-patch]

Summary

Add MiniMax Cloud TTS as a new dubbing provider for VideoLingo, complementing the existing TTS options (Azure, OpenAI, Fish, Edge, etc.).

What's included

• New TTS provider (core/tts_backend/minimax_tts.py): Supports speech-2.8-hd (recommended, HD quality) and speech-2.8-turbo (faster) models via the MiniMax T2A v2 API
• 12 built-in voices: 5 English voices + 7 multilingual voices, selectable from the Streamlit sidebar
• Streamlit UI integration: Voice selector with human-readable labels, model selector, and API key input in the Dubbing Settings panel
• Full i18n support: Translation keys added for all 7 locales (en, zh-CN, zh-HK, ja, es, fr, ru)
• Documentation: MiniMax TTS added to the TTS comparison tables in both English and Chinese docs
• Config: minimax_tts section added to config.yaml with sensible defaults

How it works

MiniMax TTS uses a simple REST API (OpenAI-compatible auth pattern). The provider:

1. Sends text to https://api.minimax.io/v1/t2a_v2
2. Receives hex-encoded MP3 audio
3. Converts to WAV using pydub (already a project dependency)

Users just need a MiniMax API key from minimax.io — MiniMax is already listed as a recommended LLM provider in VideoLingo's docs.

Files changed (15 files, ~500 additions)

File	Change
core/tts_backend/minimax_tts.py	New TTS provider
core/tts_backend/tts_main.py	Router entry for minimax_tts
core/st_utils/sidebar_setting.py	UI controls for voice/model selection
config.yaml	Default config section
docs/pages/docs/start.en-US.md	EN docs update
docs/pages/docs/start.zh-CN.md	ZH docs update
translations/*.json (7 files)	i18n keys
tests/test_minimax_tts.py	8 unit + 3 integration tests

Test plan

• 8 unit tests pass (mocked API, voice validation, parameter checking, directory creation)
• 3 integration tests pass with real MiniMax API (English text, Chinese text, turbo model)
• Generated WAV files are valid audio with reasonable duration
• Manual test: select minimax_tts in Streamlit sidebar, configure API key, generate dubbed audio