[feature] TrimPrimers can trim only R1s

[nh13]

Alternative to #679. The advantage of this PR is it's a bit smaller of a modification. The negative is using -1 as a proxy for None. In >90% of cases I'd prefer to use Some/None, but I think this is simple enough of a change to make it work with TrimPrimers

[FelixMoelder]

Thanks for offering an alternative variant of single primer trimming.
@dlaehnemann has created a small set of test data which we applied on your and our implementation (see #679).
Performing a diff on both results showed that reads where an amplicon was found are trimmed identical by both implementations.
In case no amplicon can be found our implementation does some weired stuff trimming both ends of a read by the maximum primer length.
We did not investigate this further as it is obviously a bug which does not occur in your implementation.

Regardless on comparing both results we also checked if the reads are trimmed as we would except (and double checked the more complex cases).
It looks like everything is fine so far and the more complex cases (leading softclips, extremely short reads, reads w/o amplicon) are also handled correctly.

Therefore, we suggest merging your implementation as it is also much cleaner and we don't see a reason trying to fix ours.

Test_primers.zip

[nh13]

@FelixMoelder thank-you for testing this implementation, we'll get it code reviewed and merged!

[FelixMoelder]

@nh13 Can you estimate when this will be merged?