Refine minimal QC modules and reporting

[SamBryce-Smith]

RSeQC has some useful modules but is super super slow (looking at you geneBody_coverage.py). Some we have in there currently are probably also not so useful and are clogging up the MultiQC report.

RNA-SeQC - https://github.com/getzlab/rnaseqc - has a lot of the same functionality but is much faster. Mainly would like to use this for strandedness reporting & gene body coverage profiles but there should be other useful metrics in there. https://github.com/getzlab/rnaseqc/blob/master/Metrics.md

I also liked the look of Picard's collectRNASeqMetrics - https://broadinstitute.github.io/picard/picard-metric-definitions.html#RnaSeqMetrics - but have to decide how much of this is already covered by other tools and whether this is necessary.

I started working on this on branch refactor_units_qc, but we should come to a decision as to what to include/remove. Annoyingly all of these tools require different format annotation files (GTFs/BED12/refFlat etc.) so need to do some interconversion.

Some other possible additions - https://github.com/zavolanlab/tin-score-calculation - handy to know if seeing extensive RNA degradation within samples. We'd have to plug this into multiQC ourselves, though this has already been done in their pipeline https://github.com/zavolanlab/zarp