Add CreateChatCompletionStreamResponse as an Event-stream Response Type to the /Chat/completions Endpoint

[paulhdk]

In the /chat/completions endpoint definition under x-oaiMeta it says:

x-oaiMeta:
    name: Create chat completion
    group: chat
    returns: |
        Returns a [chat completion](/docs/api-reference/chat/object) object, or a streamed sequence of [chat completion chunk](/docs/api-reference/chat/streaming) objects if the request is streamed.

The "streamed sequence of chat completion chunk objects" refers, I believe, to the CreateChatCompletionStreamResponse type, which is defined, but never explicitly linked to the createChatCompletionRequest() function.

Since it is never defined that createChatCompletionRequest() may return a stream of chunk objects, tools such as the swift-openapi-generator package, which I have been using to implement the OpenAI OpenAPI spec, cannot generate the correct code for the spec.
For more context, see the corresponding issue in the swift-openapi-generator repo.

This PR links the CreateChatCompletionStreamResponse type to the createChatCompletion() function.

[PSchmiedmayer]

Thank you @paulhdk for creating the PR; it would be amazing to see this merged to allow generators to pick up on this information and correctly parse steaming requests.

[PSchmiedmayer]

@kwhinnery-openai Wondering why this PR was closed; would it make sense to add this to the specification to properly parse streams based on the spec? It would be great to have this in here to avoid that we need to maintain a copy of the spec of our client stubs. Thank you for the support and maintaining this repo!

CC: @paulhdk.