feat(l2g): limit l2g predictions to gwas-derived associations

[ireneisdoomed]

... as discussed with @d0choa and @addramir

[codecov-commenter]

Codecov Report

Attention: 129 lines in your changes are missing coverage. Please review.

Comparison is base (42b366c) 85.67% compared to head (bdcb218) 86.20%.
Report is 54 commits behind head on dev.

Additional details and impacted files

@@            Coverage Diff             @@
##              dev     #408      +/-   ##
==========================================
+ Coverage   85.67%   86.20%   +0.53%     
==========================================
  Files          89       95       +6     
  Lines        2101     2596     +495     
==========================================
+ Hits         1800     2238     +438     
- Misses        301      358      +57     

Files	Coverage Δ
src/airflow/dags/common_airflow.py	90.38% <100.00%> (ø)
src/airflow/dags/dag_preprocess.py	100.00% <ø> (ø)
src/airflow/dags/finngen_preprocess.py	100.00% <100.00%> (ø)
src/airflow/dags/gwas_catalog_harmonisation.py	43.47% <ø> (ø)
src/airflow/dags/gwas_curation_update.py	100.00% <100.00%> (ø)
src/otg/cli.py	91.66% <100.00%> (+21.66%)	⬆️
src/otg/common/session.py	87.50% <100.00%> (+0.32%)	⬆️
src/otg/config.py	100.00% <100.00%> (ø)
src/otg/dataset/dataset.py	91.80% <100.00%> (ø)
src/otg/dataset/l2g_feature_matrix.py	82.92% <100.00%> (+7.31%)	⬆️
... and 35 more

[d0choa]

Sounds good. Look at the comment for a potentially useful refactor

[d0choa]

this could be a reasonable StudyLocus method. Most likely we will use it again study_locus.filterByStudyType(study_index, ["gwas"]).

It's a refactor so it could go in a different PR