sc-seq diagnostic plot improvements

[erflynn]

• [ ] files are sometimes too large to view in RStudio. Make these smaller -- use rastering or a .png or something else
• [ ] make log-scale plots for nCount and nFeature as well
• [ ] consider adding automated thresholding (e.g. otsu, multiotsu, li) + plots for this?
• [ ] duplication across cell-filter and automated-processing
• [ ] option to add downstream steps to the cell-filter up to find markers -- this could be helpful for diagnostic purposes if we do not get many singlets from demultiplexing
• [ ] pull in CR nUMI cutoff, est number of cells for comparison
• [ ] consider adding dropkick
• [ ] ADT cutoffs still written for scRNA