feat: enable dynamic embedding models #1060

[e7217]

This PR is related to issue #1060 .
I have tried to make it easier for users to apply embedding models that are not included in the embedding models list.

Below is how to apply the embedding model in the YAML file:

vectordb:
- name: autorag_test
  db_type: milvus
  embedding_model: 
  - type: huggingface
    model_name: intfloat/multilingual-e5-large-instruct
  ...

or

vectordb:
- name: autorag_test
  db_type: milvus
  embedding_model: 
  - type: openai
    model_name: text-embedding-3-small
  ...

However, the previous method can still be used:

vectordb:
- name: autorag_test
  db_type: milvus
  embedding_model: openai_embed_3_small
  ...

I have checked that this change does not impact the existing functionality.
I appreciate your review. Thank you

[vkehfdl1]

LGTM
@e7217 Thanks for the contribution.
I changed a bit and add documentation.
Overall it is great feature for everyone.
We are late for the review and sorry for that.
Happy new year!!