Run jobs parallel again (#47)

When running an ensemble, one of the changes in [Cleanup of requirements
and increase code
quality](45ef591),
made the jobs being run sequentially instead of parallel on the login
nodes. This MR reverts the relevant change.

Fix member selection (#46)

In case several member selections were run at the same time (as can
happen on BB), all member selections were using the same name for the
temporary file `"random_tolerances.csv"` leading to a race condition.
This should be solved by changing the name of the file to
`"random_tolerances_${experiment_name}.csv"`.

Cleanup of requirements and increase code quality (#37)

This PR cleans up multiple things

- update miniconda scripts
- merge run and development requirements
- move jenkins testing to Balfrin
- activate pylint (cleanup errors and most warnings)