Add function to get text embedding and similarity

[cthoyt]

motivated by cthoyt/ols-client#9, this adds a function for getting an embedding based on putting an entity's name + description into sentence_transformers. It also adds a quick function for calculating term similarity based on text embeddings

import pyobo

>>> pyobo.get_text_embedding("GO:0000001")
[-5.68335280e-02  7.96175096e-03 -3.36112119e-02  2.34440481e-03 ... ]

>>> pyobo.get_text_embedding_similarity("GO:0000001", "GO:0000004")
0.24702128767967224

future:

1. batch mode
2. incorporate synonyms