feat: allow the specification of a model for fine tuned completion

[ColinEberhardt]

When using a fine-tuned model, you specify a model rather than an engine - see:

https://beta.openai.com/docs/guides/fine-tuning/use-a-fine-tuned-model

This adds support for a model parameter, e.g.

  const gptResponse = await openai.complete({
    model: "curie:MODEL-ID",
    prompt
  });

[davut]

+1

[pdhoward]

++1 please

[Lisennk]

+1

[bcazur]

+1

[dasheck0]

There were some other changes as well. I covered them in this PR #57