Support loading of wide-format phenotypic CSVs?

[alyssadai]

The current long-format digest schema for phenotypic CSVs is a lot less constrained than the imaging digest schema, as we want phenotypic dashboard views to be flexible enough to support files containing either availability info (which might be statuses) or actual data values.

A long-format schema was initially picked for the phenotypic digest to be able to reuse methods to process the input CSV for the dashboard from the imaging digest workflow.

However, whereas we would expect imaging digest files to (in most cases) be generated directly in long format based on the dashboard schema, a phenotypic digest by contrast will most likely be some pre-existing wide tabular file that the user reshapes to long format to use as a digest. This second step is clunky and currently a bit redundant, as the dashboard will then internally transform the data back to wide format to display as a datatable.

Overall, it's likely that the long-format schema for the phenotypic digest is still more flexible and more future-proof in case we need to add more complex phenotypic dashboard features in the future. That said, we should revisit this if the current schema becomes too much of a barrier for users.