[Feature Request] Embed doesn't support Sparse Vectors?

[qhenkart]

What motivated you to submit this feature request?
I wanted to leverage the pinecone inference client to create a sparse vector

Describe the solution you'd like
I would like to leverage the pinecone inference client to create a sparse vector

Describe alternatives you've considered
Using another API and service other than Pinecone

Additional context
Following this guide https://docs.pinecone.io/guides/data/encode-sparse-vectors I was going to use the inference.embed functionality to create sparse vectors for a hybrid search approach. However when trying to implement it, it became quickly obvious that the SDK only supports dense vectors, as it doesn't support any sparse vector output in the API response. This forces me to use cURL or look for another option that has dedicated hybrid search or keyword support.

Is there a way to perform this operation via the Go SDK?