clarifications

[lwaldron]

@krlong68 can you add these clarifications?

1. UUID to Sample ID Mapping: The repository doesn't explain where the mapping between UUIDs and sample IDs is stored or how it's provided to generate_sql.py.
2. DuckDB Execution Command: The README states that generate_sql.py produces .sql scripts and mentions a Singularity .def file, but it lacks the exact command needed to actually run the SQL scripts with DuckDB. It would be helpful to include an example command (e.g., duckdb -c ".read output/run_name.sql" or similar).
3. Parquet Destination: The documentation doesn't explain where DuckDB writes the final .parquet files (e.g., whether they are written locally or pushed back to a cloud storage bucket) and how they are subsequently uploaded to Hugging Face or Zenodo for the R package to consume.

[krlong68]

Note on 1: I will describe how IDs are used in this repo / provided to generate_sql.py. I'll add the actual UUID mappings and the code used to create them in parkinsonsManualCuration, since that happens before the pipeline is run.