Split accession ids list in est.nf `split_sequence_ids` process block

[rbdavid]

Just to note, this is likely where the uneven distribution of sequence lengths (resulting in uneven axa blast run times are being seen) is being introduced. Is there a way to split accession ids with knowledge about the associated seq length?

Originally posted by @rbdavid in #293