Read structure for CELseq protocol swaps R1 and R2

[mcmero]

The scPipe::get_read_str('CEL-Seq') returns list(bs1 = -1, bl1 = 0, bs2 = 6, bl2 = 8, us = 0, ul = 6), which indicates that R2 contains the barcode information. However, typically the barcode information is in R1 for CEL-Seq. This should also be the case for CEL-Seq2 and 10X protocols. The us and ul parameters also seem to assume that the UMI is in R2 (rather than R1), so I was only able to process some CEL-Seq2 data by swapping the R1 and R2 fastq input files. It would be great if this could be clarified in the documentation or the read structure swapped to reflect these protocols.