Dependent on #277. Related to #106, #211, and #295.
Since we now store traceback information for failed ProtocolDAGResults in the state store to enable Task restart policies, it would be valuable to users to be able to easily and quickly introspect the tracebacks that failed Tasks exhibit.
We propose the following new methods:
get_task_tracebacks: for the given Task ScopedKey, return a list of dicts, with ProtocolUnitFailure GufeKeys as keys, and their tracebacks as values.