The application author creates a JupyterLab build on their machine using the
core JupyterLab application. They can then serve their files by subclassing the
LabServerApp with the appropriate configuration and creating a Python entry
point that launches the app.