Cross-script references

[piotor87]

core uses data and config from finngen_qc and this should/could be improved. E.g. harmonization units are used to validate the extracted outcomes and the same mappings are used to clean up the units to begin with.

Ideally one could make the pipeline run all at once, but it's still good to keep things seperate, as:

1. we check for duplicates after omop mapping
2. there's no need to start from raw data when OMOP mapped data has been generated

It's good to remember about this and decide what to do later as potentially more feature pop up.

[piotor87]

Ideallly this should be solved in V4 IF/when we integrate extracted data into the harmonization pipeline

[piotor87]

SOLUTION: One single script