Improved documentation, validation and completeness summarization

[yavyx]

What's new

• Added instructions to install the clinical_etl package into the virtual environment (if needed). DIG-1613
• If a treatment start or end date is earlier than date of diagnosis, return a validation warning instead of an error.
• Added completeness_table.py script written by @kcranston to generate a readable CSV file of the required field completeness stats.

How to test

• pip uninstall clinical_etl from virtual environment, confirm that CSVConvert can't find the module, then reinstall with pip install -e clinical_etl_code/ and confirm it does work.
• Create or validate a dataset with treatment start and end dates before date of diagnosis and check the warnings.
• Run completeness_table.py with any clinical _map.json file and check the output CSV file.

[daisieh]

Looks good, everything works as expected.

[mshadbolt]

everything works as expected for me too! Thanks!