[Feature] Support audio modality via Chat resource

[MaestroError]

Background

I was trying to implement audio modality with Chat completions API:

Into the LarAgent and it returns empty message:

[choices] => Array
        (
            [0] => OpenAI\Responses\Chat\CreateResponseChoice Object
                (
                    [index] => 0
                    [message] => OpenAI\Responses\Chat\CreateResponseMessage Object
                        (
                            [role] => assistant
                            [content] =>
                            [toolCalls] => Array
                                (
                                )

                            [functionCall] =>
                        )

                    [finishReason] => stop
                )

        )

I tried to debug it and found out that API actually returns the audio content but CreateResponseChoice doesn't stores it, I printed the response:

client/src/Resources/Chat.php

Line 33 in 2d305c8

$response = $this->transporter->requestObject($payload);

[choices] => Array
                (
                    [0] => Array
                        (
                            [index] => 0
                            [message] => Array
                                (
                                    [role] => assistant
                                    [content] =>
                                    [refusal] =>
                                    [audio] => Array
                                        (
                                            [id] => audio_686ad073bba08191bd03d9c9541b0f58
                                            [data] => SUQzBAAAAAAAI1RTU0UAAAAPAAADTGF2ZjYwLjE2LjEwMAAAAAAAAAAAAAAA//OEwAAAAAAAAAAAAFhpbmcAAAAPAAAApQAAjcAABgoLDhEUGBscHyQlKS0uMjU3Ojw.... (Long Base64 string)
                                            [expires_at] => 1751834243
                                            [transcript] => Your name is John. Do you have any other questions for me, John?
                                        )

                                    [annotations] => Array
                                        (
                                        )

                                )

                            [finish_reason] => stop
                        )

                )

I would like to add support for Audio modality output via Chat resource. Learn more here: https://platform.openai.com/docs/api-reference/chat/create

[iBotPeaches]

Able to meet me halfway with a sample to reproduce this easily? I'll add it to my collection and fix on Monday - https://github.com/iBotPeaches/openai-php-laravel-test/blob/master/app/Console/Commands/ChatStreamIncludeUsageTest.php

[MaestroError]

@iBotPeaches Try this:

 $stream = OpenAI::chat()->create([
            'model' => 'gpt-4o-audio-preview-2025-06-03',
            'modalities' => ['text', 'audio'],
            'audio' => ['format' => 'mp3', 'voice' => 'nova'],
            'temperature' => 0.7,
            'messages' => [
                ['role' => 'user', 'content' => 'What is the capital of France?'],
            ],
        ]);

[iBotPeaches]

Made a sample to prove it works - iBotPeaches/openai-php-laravel-test@bc53d4d
PR - #629

[MaestroError]

@iBotPeaches Thank you 🙏