Add support for maxCompletionTokens and reasoningTokens in openai_dart

[davidmigloz]

maxTokens is deprecated in favor of maxCompletionTokens, and is not compatible with o1 series models.

Before:

final res = await client.createChatCompletion(
  request: CreateChatCompletionRequest(
    model: ChatCompletionModel.modelId('gpt-4o'),
    messages: [
      ChatCompletionMessage.system(
        content: 'You are a helpful assistant.',
      ),
      ChatCompletionMessage.user(
        content: ChatCompletionUserMessageContent.string('Hello!'),
      ),
    ],
    maxTokens: 2,
  ),
);

Now:

final res = await client.createChatCompletion(
  request: CreateChatCompletionRequest(
    model: ChatCompletionModel.modelId('o1-mini'),
    messages: [
      ChatCompletionMessage.system(
        content: 'You are a helpful assistant.',
      ),
      ChatCompletionMessage.user(
        content: ChatCompletionUserMessageContent.string('Hello!'),
      ),
    ],
    maxCompletionTokens: 2,
  ),
);

If you are using the o1 model, you can now see how many reasoning tokens it has consumed in the usage object:

final res = await client.createChatCompletion(...);
final reasoningTokens = res.usage?.completionTokensDetails?.reasoningTokens;