show is thread locking on Jupyter Notebook

[mmhobi7]

Lenspy seems to be thread locking on Jupyter Notebook.

Side note: There seems to be no way to add multiple figures to the Flask container, or spin new containers without manually doing so.

[serant]

This is a current limitation of lenspy, however it can be improved on by implementing the following:

Preamble: Ideally, lenspy would be integrated into Plotly by utilizing the underlying Javascript events, however this would require modification of the source for plotly-python's codgen to expose events to the Python API.

Instead the following is proposed (I will be working on an implementation similar to the below):

There are three possible environments that a Figure can be deployed in:

1. Offline plot mode (using plotly.express or plotly.graph_objects directly) (in this environment, we will deploy a Dash server to grant access to the events API provided by Dash to implement lenspy)
2. Online as part of a Dash app, where a graph server and access to events is possible (therefore we should be using the existing server and Dash's events API to integrate lenspy)
3. As part of a Jupyter where we should be using FigureWidget to access the event loop of the plot (since FigureWidget utilizes ipywidgets under the hood)