Expands simulate pheno with query.json and metadata files

[mcamarad]

This PR does the following

• Allows to simulate data using query.json and metadata
• It adds a testing profile for generating CB output data

Details:

• Adds script to generate config files (generate_sim_config.R)
• Refactors simulate_cb_output.R to allow new features
• Adds pheno_col with fraction_of_cases, type, name and case_group as params in config file
• Extracts distribution params from query.json and sets default sampling rules for numerical variables ('normal distribution')
• Parametrizes all options related to pheno_col as well as query file, metadata file.
• Enforces numerical data into positive values
• Removes the possibility of adding NAs for now
• Updates docs

Comments:

• Requires to query filters from interest columns in order to generate data

CI test

https://github.com/lifebit-ai/simulate/actions/runs/397349967

Test on cloudOS

https://cloudos.lifebit.ai/app/jobs/5fc8aab2cc96a20112a87aa5