Select SEQ_MAX_LENGTH based on data

[haghshenas]

At the moment SEQ_MAX_LENGTH is hard coded! (50000)
This will cause problems for longer reads, for example for Nanopore data.
Choose SEQ_MAX_LENGTH based on the longest read in each chunk.