[Feature Request]: Add whisper large v3 turbo

[thewh1teagle]

Describe the feature

Recently OpenAI released new v3 turbo model which works 8x faster! that's amazing news. the current medium model works just 2x faster. so it's 6x faster with better accuracy!

openai/whisper#2361

We would like to prepare GGML for using in Vibe as soon as possible

https://github.com/thewh1teagle/vibe/blob/main/docs/MODELS.md#prepare-your-own-models