Feature Request: facet_row support in px.imshow

It would be very convenient to be able to utilise the facet_row kwarg with plotly.express.imshow.

Current Behaviour

Currently, px.imshow only takes the facet_col kwarg, limiting the display of multiple heatmap-like traces to a single dimension (e.g. layer slices through a 3D CT image). It is not possible to visualize 4D data without utilising the animation_frame kwarg.

Motivation

In a number of scientific and engineering use-cases, it is useful to visualize data which possesses a dense 4D tensor structure. Any experiment using full factorial design with at least four factors, common in machine learning and biological science applications, falls into this category. Providing a better convenience function for generating heatmap-style plots would improve the utility of Plotly for publication-standard multivariate data visualization. It should be noted that whilst parallel category and parallel coordinate plots offer an arguably better visualization tool for this sort of data, they are still underutilised in publications, and, as such, an expanded imshow still has good utility.