The /api/flamenco/managers/{manager-id}/depsgraph endpoint returns the list
of runnable tasks belonging to runnable jobs, assigned to that manager. All
tasks that were "queued" are set to "claimed-by-manager".

Furthermore, upon updates it is also checked whether this worker is allowed
to update the task at all. If not, it is sent a HTTP status 409 Conflict.

These make mocked HTTP requests to the Manager itself easier to handle.

Failure threshold is set to 10%, so if 10% of the tasks fail the job fails.
At this moment there are jobs with 15-20 tasks, and this threshold will
allow 1 task to fail before failing the job.

If there are actual updates to push, still logs at info level.

This redirects to /flamenco/{project-url}/jobs/{job-id}, so that the
Blender Cloud add-on can open a webbrowser to show a newly created job,
without having to query for the project URL too.

The code was copied from Attract, and doesn't make sense in Flamenco;
those activities don't relate to Flamenco at all, since we don't have
any activity logging for jobs/tasks, nor comments.

Also removed superfluous \n in log messages.

By default logs only WARNING and higher. I've added -debug and -verbose
flags for respectively DEBUG and INFO level logging.

Most logger calls are still just INFO-level, will be migrated in a future
commit.

Thanks to Dr. Sybren!

This prevents diversion between version numbers.

This allows the manager to re-queue the worker's current task.

This allows for more complex Blender commands. Before this commit, the
setting could only point to an executable; now it is also possible to
provide extra CLI arguments to Blender.

The command runner just lets the task runner know it failed, which is then
forwarded to the Worker class. It's that class' decision how to handle that
with regard to task status.

Any active task assigned to the worker will be reset to status
'claimed-by-manager', so that it can be re-executed by another worker.

Plus minor tweaks and icons for action buttons