Dataset lifecycle scenario tests

[benkamphaus]

This is essentially a recovery/remembering of the PICI scenario tests, for ensuring proper cleanup, install after retract, diffing, etc. for a sequence of operations on datasets (eg prepare, transact, prepare with change, diff, transact diff, retract, transact second clean, dataset looks the same, etc). This will need to be done with new data and resources since the old ones are no longer available. This is an important component of ensuring the dataset ops are robust enough to be relied on for non-alpha use cases.

[benkamphaus]

Currently in limbo while assessing the importance of diff/retract, as opposed to expecting multiple datasets and versions to be transacted separately.