Refactor coSMicQC notebook

[jenna-tomkinson]

Seeing how this plays out I noticed it might be possible to:

• Define a yaml file like the one found within cosmicqc here (which includes your threshold sets with the names you've specified manually).
• Use that yaml file to run cosmicqc.label_outliers(..., feature_thresholds_file=<yaml_file>, ...)
• Export the file as you need below (also skipping the step to populate the data manually below).

Doing this might lower the complexity involved with the code more generally. It'd also prepare for the threshold sets to be added to cosmicqc :) .

Originally posted by @d33bs in #39 (comment)