pi-multix 是做什么的？
AI multimodal generation and media processing through the multix tools (multix_image, multix_video, multix_audio, multix_media, multix_doc, multix_check, multix_run). Use when the user wants to generate or edit images (text-to-image, image-to-image, i2i, style transfer, watercolor, cyberpunk, reference image, Nano Banana, Flux, Seedream, Imagen, gpt-image), generate video (text-to-video, image-to-video, i2v, Veo, Hailuo, Kling, Seedance, LTX, Motion), produce speech or transcribe it (TTS, STT, s