Consider defaulting flat prior across genotypes

[timothymillar]

Currently we default to a Bayesian approach informed by an expected inbreeding coefficient of zero. This works well if accurate inbreeding priors are specified but we can expect many people will ignore this parameter. With an inbreeding prior of zero we are implicitly favoring heterozygous genotypes (especially when there are many alleles). It may be better to default to a truly "neutral" prior in which all genotypes are equally likely (analogous to maximum likelihood). This can be achieved by calculating the prior probability of any genotype as 1/G where G is the total number of possible genotypes.