remove ANI-constraint on uploading MAGs

[liamfitzstevens]

Your documentation reads as follows: “There should only be one MAG submitted for each species within a biome.” In human gut metagenomics, this would mean MAGs need to be dereplicated to 95% ANI per metagenome; that every species present in a given human gut can be represented by only 1 MAG. Wouldn't it make more sense for users to upload MAGs dereplicated to whatever % ANI was used in their study? For example, in the case of PRJEB46788, we used MAGs dereplicated at 99.9% (i.e., there can be multiple MAGs of the same species assembled from the same metagenome): uploading MAGs dereplicated at 95% -as recommended in your current documentation- would be useless for people looking to replicate our findings or use the MAGs for their own studies, because we would have only uploaded a subset of the total MAGs used in our study.