write script to aggregate results from replicates in a single file

[petrelharp]

Right now the compute_stats.py script puts information in three files:

• repname.json (recapitation parameters)
• repname.stats.csv (last column is pi, I think)
• repname.pairstats.csv (has dxy and Fst, we want dxy)

We also need to pull the values in params.json. All these things from each replicate need to go in a single line of a single file, for abc; I propose writing a script which takes a directory name on the command line and pulls the information from all the replicates out of there.