New docs section: configuring nextflow and running on supercomputers.

[darcyabjones]

We should have a bit of information about how to specify different profiles, and how you can setup your own config file to run in a custom way.

E.g. People could optimise it to run on their local super-computer by setting the job-scheduler, specifying how many CPUs a task can use per node etc.