Experiments table: action per item to run experiments

[shcheklein]

Originated at #1120 (comment)

Single new experiment

A single experiment to queue, rather than how to submit a whole batch of experiments. For example, there could be a GUI that shows all the parameters, and users can edit their values and click "queue experiment" without needing to edit yaml files or run terminal commands.

We should be able to run experiments based on any of the experiments in the table or workspace.

Batch

Concerns:

As @daavoo alluded, when it comes to larger scale hyperparameter tuning, users probably can't avoid doing this programmatically since there are so many experiments to queue, the methods of defining the hyperparameter space are too varied, and frequently they can't even all be queued up front because later iterations base their values on the results of previous iterations.

I think it makes sense for advanced scenarios and especially when experiments are expensive to run (deep learning), in case of "classical" DS random forest, xgboost it's easier to run a small grid batch .

I would add some basic way to specify a list of values at least and/or simple start-step-end UI.

[rogermparent]

While #1267 is slightly more generic than this Issue, they're so close that it's probably best to merge the two. I've already added the info into the other ticket, so this one can safely be closed.