Create a job array script rather than separate job script for each experiment

[prehensilecode]

Currently, split_nlogo_experiment.py creates a separate job script for each experiment. For a large number of experiments, this can be cumbersome.

Instead of separate job scripts per experiment, it should produce a single job script defining a job array. Job arrays are supported by PBS/Torque, Slurm, and Grid Engine.

An example in Slurm:

#SBATCH --job-name=my_job
#SBATCH --array=1-400