Will n-D observation states be supported?

From the util function (and testing with `ALE/Bowling-v5`), it appears that only 1-D (vector) observation shapes are supported. Allowing for n-D shapes can be interesting as it can open up experiments things like text-embeddings, while maintaining the given structure.

Is there a plan to add this in?

Start by locating the util function referenced in the report and reproduce the observation-shape behavior with ALE/Bowling-v5. Determine how n-D observation states should preserve their given structure, then confirm that the intended text-embedding experiments are supported.

Written by the indexing model from the issue text.