Clean up driver_internal_info after tear_down

[Lucas Alvares Gomes]

After the node is successfully teared down, the driver_internal_info still populated with information from the previous instance. Things like the root UUID of the previous instance image filesystem or whether the previous image filesystem was a full disk image or not.

| created_at | 2015-05-05T14:25:49+00:00 |
| driver_internal_info | {u'clean_steps': None, u'root_uuid_or_disk_id': u'ebab4e22-91c1-4a0d- |
| | 8e1e-87935dfb6891', u'is_whole_disk_image': False} |
| chassis_uuid | 63360398-e899-4812-87ee-d2b9baf6fd65 |
| instance_info | {} |
+------------------------+------------------------------------------------------------------------+

I think those should be cleaned since they get exposed to the Ironic user

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/180220

[OpenStack Infra (hudson-openstack)]

Change abandoned by Dmitry Tantsur (<email address hidden>) on branch: master
Review: https://review.openstack.org/180220
Reason: Hi! Judging by the bug status, this patch is abandoned. Please feel free to resurrect it if you plan to continue

[Julia Kreger (juliaashleykreger)]

This code pattern was largely revised, and we remove these field contents upon the completion of teardown. We don't purge *everything* but everything pertinent in that part of the workflow. As such, this bug no longer seems valid since we also cannot just entirely delete the entire field every single time as we store state related data for the driver's current step or sequence of operation in the same field. As such, marking invalid at this time.