Post mapping length filter

[jfy133]

nf-core/eager pull request

This is to close #522 . This allows a user to turn off length filtering at adapteremoval and instead run length filtering on mapped (only!) BAM files.

Therefore all reads in a library should in principle go into mapping after going through AdapterRemoval (and after merging to account for pairs, and assuming the vast majority of read pairs overlap), and thus will give a more realistic endogenous DNA value for 'economic' reasons.

PR checklist

• This comment contains a description of changes (with reason)
• If you've fixed a bug or added code that should be tested, add tests!
• If necessary, also make a PR on the nf-core/eager branch on the nf-core/test-datasets repo
• Ensure the test suite passes (nextflow run . -profile test,docker --paired_end).
• Make sure your code lints (nf-core lint .).
• Documentation in docs is updated
• CHANGELOG.md is updated
• README.md is updated

Learn more about contributing: CONTRIBUTING.md

[stschiff]

Definitely a very elegant addition, and I like the explanations in the help messages! If we go that route for improving the endogenous DNA calculation (which I'm not yet completely convinced of, see my comments to #522), we should clearly advertise it to users, especially for single-stranded library prep and very short reads. As this may become more and more common, we need to think whether it's safe to even recommend that as the default.