Draft: Dirty fix for output grid orientation

... and daily time step.

This is only a quick, intermediate fix to test the principle working of coupling with Formind.

It seems like mHM provides swapped nrow and ncol variables. The grid definition in the test domain is as follows:

ncols         6
nrows         9
xllcorner     3973369.0
yllcorner     2735847.0
cellsize      24000.0
NODATA_value  -9999

(from header.txt associated to pre.nc)

while mHM reports ncols=9, nrows=6.