starting conditional analysis as a new file

[annacuomo]

Super work in progress attempt at implementing a conditional analysis part of the pipeline.

Ideally, I'd like this script to:

• take in results from round 1 of running the pipeline (no SNPs in the covariates)
• extract all genes with a significant SNP (at a given threshold) and retains info on the top SNP
• run step2 of the pipeline again (no need to rerun step1 as it is the same) this time conditioning on that top SNP
• at the end summarise results, extract genes with a(nother) significant SNP (from round2)
• runs step2 only for those conditioning on both SNP1 and SNP2 (saved in a dict specific to that gene)
• continues (round3, etc) until no genes have any significant SNP

@Alexander-Stuckey @MattWellie (@katiedelange, @hopedisastro) can I pick your brains on how to best do this? See my starting point in this PR. I opted for doing this as a separate script because it was getting quite messy? This pipeline seems to similarly run the conditional analysis after the main one (https://github.com/populationgenomics/tob-wgs/blob/main/scripts/eqtl_hail_batch/launch_eqtl_spearman.py#L829)

Also I am trying to have the option to run both a common variant (single-variant test) and a rare variant (set-based test) version in one, but open to making this two / split the code more