feat: add MiniMax as LLM and TTS provider

[octo-patch]

Summary

Add MiniMax as both an LLM and TTS provider for ShortGPT.

Changes

LLM Provider (Chat Model)

• Add MiniMax as alternative LLM provider in gpt_utils.py via OpenAI-compatible API
• Support MiniMax-M2.7 and MiniMax-M2.7-highspeed models (204K context window)
• Priority order: Gemini > MiniMax > OpenAI (backwards compatible)
• Temperature clamping for MiniMax API constraints (0.01-1.0)
• Fix num_tokens_from_messages() to gracefully handle non-OpenAI models

TTS Provider (Text-to-Speech)

• Add MiniMaxVoiceModule extending VoiceModule ABC
• Support speech-2.8-hd model with 12 built-in voices
• Integrate into all 3 UI tabs: Short Automation, Video Automation, Video Translation
• Add MiniMax TTS as selectable TTS engine alongside ElevenLabs and EdgeTTS

Configuration

• Add MINIMAX to ApiProvider enum
• Add MINIMAX_API_KEY input field in Config tab UI

Tests

• 24 unit tests (LLM provider selection, temperature clamping, TTS module, API requests)
• 6 integration tests (LLM chat completion, streaming, TTS audio generation)

Files Changed (13 files, 656 additions)

API Documentation

• Chat (OpenAI Compatible): https://platform.minimax.io/docs/api-reference/text-openai-api
• TTS: https://platform.minimax.io/docs/api-reference/speech-t2a-http

[vercel]

Someone is attempting to deploy a commit to the rayventura's projects Team on Vercel.

A member of the Team first needs to authorize it.