Add new job_details.model key to transcribe return dict

[ururk]

Per discussion #1112

I've added a new key to the return dictionary from the transcribe def:

{
    "job_details": {
        "model": "base"
    },
    "text": " And so my fellow Americans, ask not what your country can do for you, ask what you can do for your country.",
    "segments": [{"id": 0, "seek": 0, "start": 0.0, "end": 11.0, "text": " And so my fellow Americans, ask not what your country can do for you, ask what you can do for your country.", "tokens": [50364, 400, 370, 452, 7177, 6280, 11, 1029, 406, 437, 428, 1941, 393, 360, 337, 291, 11, 1029, 437, 291, 393, 360, 337, 428, 1941, 13, 50914], "temperature": 0.0, "avg_logprob": -0.2968965598515102, "compression_ratio": 1.3544303797468353, "no_speech_prob": 0.04440217837691307}], "language": "en"
}

Besides model, there may be other parameters passed to whisper that could be useful, hence the nested key instead of a top-level property.

[ururk]

Would you prefer the model be part of the main result dict instead of a in a sub dictionary?

[ururk]

One additional comment. The model_name param could also be passed to whisper as a full path - in that event the string stored would be the full path to the model used, and not the name of the model (ie, medium.en). I'm not sure how this should be handled. This also calls into question whether it needs to generate three keys, which I'm doing in my own code (not this pull request):

model_name (if a path is given to whisper, computed from the path)
model_path (path to model, including file with extension)
model_checksum (SHA256 checksum of the model file, used to further compare runs across time)