add conflict at t-1 as sample data about copro HOT 5 CLOSED

@wande001
now possible to use conflict at t-1 for reference run (see #111).
not yet possible for projections though.

from copro.

in b380fcb, the way conflict at t-1 is determined is upgraded.

• determine neighboring polygons in neighboring matrix;

• per polygon, check if neighboring polygons experienced conflict;

• if so, add the conflict events for all neighboring polygons and log-transform it.

from copro.

in 51c2150, additional changes based on Halvards comments are implemented:

• per polygon, check whether there was conflict at t-1 (bool);

• per polygon, determine the log-transformed count of conflict events in all neighboring polygons (float).

Besides, the first year will be skipped to avoid any complications when calculating conflict at t-1 for the first time step.

from copro.

has been changed by now:

• only computing binary value whether conflict occured in neighboring polygons at t-1 because we cannot project forward actual numbers of conflict casualties, but only conflict occurrence (0/1).

from copro.

succesfully implemented in #120

from copro.