Vignettes: Code refactoring and Figure enhancements.

[enblacar]

Hi!

I am a very big fan of decoupleR and I use it as an integral part of my analyses both for single cell and bulk transcriptomics data. For this, I have been thinking for a while in ways I could help enhancing this great R package.

I noted that the vignettes could use some improvement and, if you agree with it, I would like to propose the following changes:

• All the code has been styled so that it stays within the span of the code blocks as much as possible, since code folding is not enabled in the website. Perhaps reducing the font size of the code blocks would be an option for the website.
• Added the source packages to each function, so that group_by now becomes dplyr::group_by. This helps the user get a better overview of where each function originated from.
• Added a different Red-Blue color palette to the plots that I find more aesthetic.
• Enforced squared tiles in all heatmaps so that they become perfect squares, and added white borders around for enhanced aesthetics.
• Added a black border around the cells in the volcano-like plots, to make them more aesthetic.

Suggestions for further improvement:

• Add more theming to the DimPlot and FeaturePlot figures in the single-cell vignettes, perhaps with the use of single-cell visualization packages such as SCpubr.

While I have tested the output of the vignettes with knitr, this does not incorporate further CSS theming. For this, I believe you have to test it on your end.

Let me know if this was helpful, and I am very happy to apply any other changes to this PR as per your request!

Best,
Enrique