[DOC] Pre-trained Embeddings Usage

[mvidela31]

Report needed documentation

Report needed documentation

It seems that the pre-trained embeddings feature is already supported #690. Unfortunately, there is no documentation or examples on how to use it.

Describe the documentation you'd like

It would be awesome to have pre-trained embeddings related documentation, as well as examples on how to use this feature and how to deploy a model with this feature on the Triton inference server.

Thanks for your amazing work!