Improve ls output

[mike0sv]

Now it is very simplistic and not informative. What info can we add?
Related to #153

[mike0sv]

@aguschin any thoughts?

[aguschin]

@mike0sv
I would try to keep things simple in the first iteration:

• name
• is it a model/dataset
• framework
• size

Beside this we have requirements, input/output shapes, column names, method names, artifact uri (useful if meta/artifact are located in different places). Ideally, we should add those, but not sure it's useful, the lists could be quite long.
To print those I would use simpler command, something like pprint we had. But if we expose that command, its output should be more user-friendly than simple cat. So need to think how to do that (can colorize the output to make it easier to read at least).

[mike0sv]

I guess those are all related #168 #98

[mike0sv]

also, need to catch errors on parsing objects and output them nicely

[aguschin]

related to #279

[aguschin]

ls was removed for now, closing this.