More algorithm parameters

[lisanhu]

Currently, the program allows the tuning of two parameters: seed-length and threshold ( to identify non-informative seeds).

Some more proposed parameters are listed here:

• expected error rate: This will affect cigar computation (especially for edlib). If the expected error rate is set too low, it will reduce the sensitivity, ow, it will affect the performance.
• gap length between seeds in the same iteration: This may affect the accuracy, but not quite sure the relationship.