Add support for OpenAI, Azure, Cohere, Replicate Embedding Models - using litellm

[ishaan-jaff]

This PR adds support for models from all the above mentioned providers using https://github.com/BerriAI/litellm/
TLDR

• You get support for all liteLLM supported embedding models
• I noticed you had a class for Cohere embeddings, instead of maintaining classes, liteLLM can be the simple abstraction to select your models. Simply pass the model param when calling embedding and liteLLM manages calling the correct API & translation of input + output

Here's a sample of how it's used:

from litellm import embedding

## set ENV variables
# ENV variables can be set in .env file, too. Example in .env.example
os.environ["OPENAI_API_KEY"] = "openai key"
os.environ["COHERE_API_KEY"] = "cohere key"

# openai call
response = embedding(input=[text], model="text-embedding-ada-002")

# azure call
response = embedding(input=[text], model="azure-embedding-model", azure=True)

# replicate call
response = embedding(input=[text], model="replicate/all-mpnet-base-v2:b6b7585c9640cd7a9572c6e129c9549d79c9c31f0d3fdce7baac7c67ca38f305")

[ishaan-jaff]

Cc @mishanti @FelixKirsch can you take a look at this when you get a chance ? Happy to make additonal changes/update docs based on your feedback

You can also get logging for your embedding calls made through liteLLM😊

[ishaan-jaff]

in order to use cohere/replicate/other embedding models, just pass in the the model here