Add support for Image Encoders

[mahvar]

Is your feature request related to a problem? Please describe.
I'm working on open sourcing a model that takes medical images and outputs embedding vectors. It won't fit into any of the available tasks. Please consider adding support for Image Encoders or tell me how I could do that.

Describe the solution you'd like
I'd like to see Image Encoders as a supported task type.

Describe alternatives you've considered
Considered using "Image Recognition" task, but my model would be a precursor for developing "Image Recognition" models.

Additional context
N/A