Generate front images with AI given the description in the input file

[siemdejong]

Is your feature request related to a problem? Please describe.
The user now has to give separate images and if no image is given, no image is plotted.

Describe the solution you'd like
Always show an image, regardless of the user providing one.

Describe alternatives you've considered
An AI algorithm would probably able to generate some images (the example image has been created with DALLE).

Additional context
The current Streamlit Cloud resource limits are documented here.
Any model should fit in this scenario if the model is loaded locally and no API is used.

[siemdejong]

A pro of using docker is to circumvent these limits. A bit less accessible, but model sizes are only limited by consumer hardware.