cosmoSLICS covariance matrix

[lbaumo]

1. create text file of all file names and parse filenames similar to last step
2. survey reconstruction: we do this once per realization (with seeds and LOS i think this comes to 124 realizations- double check). draw each tile randomly from a LOS (draw an integer from a vector), then use this to query the correct file path
3. create shape noise realizations- do we need to do this? compare to virginia
4. run mass maps - peak counts to create histograms
5. compute covariance matrix
6. take mean over realizations- calculate covariance matrix 1240x (once for each shape noise realization and once per each survey realization) THEN average