implement optional behaviors regarding indels

[jflot]

The option to have indels considered as missing data and not as a 5th characters is likely to be requested by many users. Then is the question of how to deal with missing data: I think that we should implement both the way Network deals with them (cf. Bandelt et al. 1999) and the more brutal way PopArt deals with them (i.e. by masking the columns that have missing data).

[jflot]

The option to mask columns containing missing data (i.e. the "PopArt way") is now implemented. There remains to implement 1) the option to consider indels as missing data and 2) the way Network deals with missing data.

[jflot]

1) is done, but 2) is not really satisfactory as we discussed (in difficult cases it means basically picking up one random allele for sites with missing data). We will need to think more about it (e.g. implementing the algorithm proposed by Joly S, Stevens MI, van Vuuren BJ (2007) Haplotype networks can be misleading in the presence of missing data. Systematic Biology 56:857–862 for minimum spanning networks?).