PNC-212: Automated docstring with sphinx (#20)

* [PNC-675] Update prod codebase with UD improvements

* [PNC-675] Update prod codebase with UD improvements

* [PNC-675] Update prod codebase with UD improvements

* [PNC-212] Automated docstring with sphinx

* PNC-212

* [PNC-212]  Automated docstring with sphinx

* [PNC-212] Clean up index pages

* PNC[212] fix GA issue with ubuntu version

* PNC[212] fix GA issue with ubuntu version

* add tests for optional imports

* undo test env patches for optional import in bert

* undo test env patches for optional import in custom hunspell

* add __module__ def to faqt top level and update index and usage pages, change theme

* Move class docstrings from init to class

* capitalize faqt name

* minor fixes

---------

Co-authored-by: Carlos Samey <carlossamey@Carloss-MacBook-Pro.local>
Co-authored-by: Suzin You <suzinyou.sy@gmail.com>
Co-authored-by: Sid Ravinutala <sid.ravi1@gmail.com>

.github/workflows/tests.yml

@@ -15,7 +15,7 @@ jobs:
       - name: Install Dependencies with Apt Get
         run: |
           sudo apt-get update
-          sudo apt-get install python3.8 jq -y
+          sudo apt-get install python3.9 jq -y
       - name: Install libraries
         run: |
           pip install '.[dev]'

.gitignore

@@ -79,6 +79,7 @@ instance/
 
 # Sphinx documentation
 docs/_build/
+docs/build/
 
 # PyBuilder
 target/

README.md

@@ -13,8 +13,7 @@ contents (e.g. FAQs) based on their relevance to the text.
 The Scorer class needs the contents and some pre-trained NLP model to do the match
 scoring. It outputs the match score for each of the content.
 
-![FAQT match scoring
-diagram](docs/readme/images/faqt_readme-how_it_works-faq_matching.drawio.png)
+![FAQT match scoring diagram](docs/readme/images/faqt_readme-how_it_works-faq_matching.drawio.png)
 
 * For Scorers that use word embeddings, scoring is based on the distances between
   words or sets of words from the message and the contents.