ML: Improve the machine learning pipeline

[jo-elimu]

Streamline the process of building, training, evaluating and deploying our machine learning models.

Storybook Reading Level Predictor

• Automate the dataset download during the data preparation step ml-storybook-reading-level#11
• Add requirements.txt ml-storybook-reading-level#8
• Add GitHub workflow ml-storybook-reading-level#9
• Add GitHub workflow for saving model as PMML
• Schedule nightly build of model ml-storybook-reading-level#12
• Add accuracy score ml-storybook-reading-level#20
• Add Mean Absolute Error (MAE) threshold check to GitHub workflow ml-storybook-reading-level#26
• Add accuracy score threshold check to GitHub workflow ml-storybook-reading-level#27
• Add nightly job for importing the latest version of the storybook reading level model webapp#1920