Interactive quality filtering functionality

Hi,

This PR adds functionality to:

• Automatically create thresholds for data via gaussian mixture models (qc_filter.automatic_thresholds)
• Plot and interactively change QC thresholds via matplotlib notebook sliders (qc_filter.quality_violin)
• Filter adata based on the estimated and adjusted QC thresholds (qc_filter.apply_qc_thresholds)

One of the main differences from before is that the thresholds are given in a dictionary of the form:

{
"column_name": {"group1": {"min": 0, "max":10},
                 "group2": {"min": 1, "max":10}},
 "another_column": {"group1": {"min": (...)},
                    (...)}
}

Which should make it easier to validate the format and use these for both obs and var interchangeably. I integrated it into the notebook, so I would be grateful for review 🙏