K-mers input to GCSA: kmer, node+position, previous characters, successive characters about vg

[ekg]

Updated title. vg should produce an output stream that can be used by GCSA2 as input. The format (as @jltsiren and I have discussed) should look like:

label, starting_position, previous_chars, successive_chars, successive_positions

Where positions in the graph are given as (node_id, offset) pairs. A label is a kmer. The previous characters are the characters immediately prior to the position in the graph--- for instance there might be only one if the position is in the middle of a node, or many if it is at the beginning and there are multiple inbound edges. The successive characters are those immediately after the kmer, and as with previous there can be several if there are several outbound edges. The same applies to the successive positions.