This is to limit clutter from old analyses.

Snakemake 5.2.2 now checks that log file names have enough wildcards.

Plotting 5UTR, CDS and 3UTR on the same graph can help discover
interesting behaviours.

Also helped overcome recursion limit issue with clustermap.

This should be easier to maintain than a manually-written list.

When low quality zones are kept for deduplication, they later need to be
trimmed.

Various modifications:
- Remap more read types
- Gather featureCounts results for further processing
- Extra chromosome sizes read from config
- Avoid impossible fold heatmaps

Large unreadable figures.

This will enable more flexibility to specify groups of libraries to be
plotted on one figure. This is currently used by metagene profiles.

Also did a little code cleaning and added a gene list to the small RNA
pipeline.

Rasterizing might make pdfs lighter.

Escaping underscore for tex-compatibility is now idempotent.

Also added a missing tex-compatibility escaping.

Normalization is by non-structural mappers.

Fixing base composition files. Use mappy instead of pyGATB.

Inactivating metaprofile around piRNA target sites due to problems with
bedtools.

Moved cleanup and backup to common library.

Optional automatic directory layout.

Made paired scatterplot plotting function part of libhts as
`plot_paired_scatters`. Modifications in small RNA-seq pipeline not
tested.

Factorized linking to raw data.

It turns out that using `include:` can fail when there are symlinks in
the path to the snakefile.