Refactor EmbeddingsResponse to Support Multiple Embeddings

[ThaKladd]

Updates the embeddings endpoint usage to handle responses containing multiple embeddings from the OpenAI API, as this is how it works as of OpenAi Embeddings docs. Previously, the implementation only supported a single embedding, and it adds network overhead when dealing with multiple chunks.

What changed:

• fromInput method on Prism::embeddings() accepts array as input
• $response->embeddings is always an array of arrays, meaning if a text is sent in, to retrieve it would be in $response->embeddings[0]

Considerations:

• I considered the possibility of adding a first() method, so that it works similar as to Laravels eloquent, but decided to follow the standard set forth by existing code

[netlify]

✅ Deploy Preview for stunning-cannoli-238aa5 ready!

Name	Link
🔨 Latest commit	0ea94e4
🔍 Latest deploy log	https://app.netlify.com/sites/stunning-cannoli-238aa5/deploys/67507eb09d4c370008afa10b
😎 Deploy Preview	https://deploy-preview-85--stunning-cannoli-238aa5.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify site configuration.

[pushpak1300]

I'm not sure if @sixlive wants to support this feature yet.
#72 (review)

6c94422

[ThaKladd]

Ah. Now I saw the comment made on the Mistral Embeddig PR. 😅 Should have looked closer.

[sixlive]

I don't know, if this is the second time being brought up, maybe I should reconsider 🤔