kernel_restart handling

as documented [here](https://app.zenhub.com/workspaces/netpyne---metacell-board-5bd615394b5806bc2bef3799/issues/metacell/netpyne-ui/213), it could happen that the kernel will restart automatically.

At the moment, inside GeppettoJupyterUtils, we are re-import the utils and re-initializing netpyne_geppetto since it's the foundation of letting the UI works, things we might want to do:

- flag that the kernel crashed and it has been restarted to the user
- save the current model and reload it once the kernel has been restarted so that the user does not have to do that manually.

Start by reading the GeppettoJupyterUtils handling and the netpyne_geppetto initialization path mentioned in the issue. The issue does not name files or tests, and leaves the desired restart notification, model save, and reload behavior open; completion criteria need to be clarified before implementation.

Written by the indexing model from the issue text.