[story] new wrapper to launch local jupyter

As a SANS instrument scientist I want to launch a jupyter instance so that I can still process data even when jupyter.sns.gov is not working.

=======================================================================

Parent	Children

Acceptance Criteria:

Description, Additional Detail, Context:

When jupyter.sns.gov is down, it is convenient to have a way to start jupyter notebooks locally on an analysis or instrument computer.

One way to achieve this is to install nb_conda_kernels in the base environment then run jupyter notebook (instructions). This will give an instance of jupyter for the current user with all kernels that contain ipykernel installed in them in the list when selecting "new" kernel.

The bottom line is to let users start running reduction with minimal efforts.

This method will be script in nsd-app-wrap, jupyter which combines these commands and launches jupyter as /usr/bin/jupyter. It will also include a .desktop file which runs jupyter notebook. The name in the application will be "jupyter local".

There should also be a .desktop file with a application name of "jupyter" which runs xdg-open https://jupyter.sns.gov.

The missing part is an additional playbook, configure_base.yml, in conda-legacy-deploy which installs nb_conda_kernels into the base environment.

WARNING

Custom widgets will not work since they need to be installed into the jupyter instance which will not be available in this mode