Refine config filters

[lindaxiang]

This is to address issue: #10
It also contains the code changes to support multi files input in PR: #4

You can test the PR using the provided test cases:

python3 prototype_mapper.py --study_id PR_12 --input_dir tests/data/source/PR_12 --output_dir tests/data/mapped/PR_12 --study_config_dir tests/studies

Test cases are under:

tests
├── data
│   └── source
│       └── PR_12
│           ├── clinical.csv
│           ├── clinical.tsv
│           └── demographics.csv
└── studies
    └── PR_12
        └── config
            ├── diagnosis.yaml
            └── participant.yaml

The test cases cover the following features:

1.  Support multi-file inputs with primary and secondary files in different formats (Gen3G use case)
2. Join the multiple input files (Gen3G use case)
3. Refine `filters` to support combined inclusion rules (BQC19 use case)
4. Add preprocess step: `calculate_field` (Gen3G use case)

[hknahal]

If I understand this correctly, this filter is supposed to only include participants who have consent_date and a diagnosis_code of U07.1 or J12.89. However, the resulting mapped files contain submitter_participant_id P006 which doesn't have a consent_date (https://github.com/Pan-Canadian-Genome-Library/pcgl-data-mapper/blob/refine_config_filters/tests/data/source/PR_12/demographics.csv). Could you clarify why this record is included?