Fix genome match

[samirelanduk]

This is the proposed fix for the genome switching issue we had. There are a few checks/changes that might be needed before merging:

• Marc could you check the changes I made to primaryclipanalysis.nf? I basically took the pattern you used to get TRIMGALORE to work and applied it everywhere a channel had two inputs that needed to be kept in line with each other.
• I’ve added a --method unique to DEDUP processes - I gather from Charlotte that this needs to be both generated from the file contents as well as being overridable, so this will need to be modified.
• I also gather there are some CLIPPY changes to be made?

[marc-jones]

Apart from the changes to subworkflows/ncrna.nf which changes the functionality, the rest looks good. I didn't anticipate the issues with having different references running simultaneously!