This version now handles a non-canonical --dir structure and directly use fastq file(s).
If no barcodeXX structure is found, NanoASV will use fastq(.gz) file(s) that are present in $DIR and still output a CSV assignment table as well as a phyloseq .Rdata formatted object. Problems linked to multiple sequence format were fixed as well