Train Spacy language model to increase the accuracy of detecting non-imperative mood

[bagashvilit]

• I have searched the issues of this repo and believe that this is not a duplicate.
• I have read the README.md and believe that my question is not covered.

Issue

CommitCanvas uses a pre-trained Spacy language model, which may not be very successful with detecting non-imperative mood. Specifically, it sometimes takes a verb in imperative mood as a noun. To resolve this issue it might be helpful to train or update the Spacy language model with commit messages. So that model is trained specifically for the commit message language.