This merge reduces the number of thread pools.

We don't need two separate executors for long-running IO threads.

Tor has a controller command, TAKEOWNERSHIP, and a configuration option,
__OwningControllerProcess, that work together to ensure Tor shuts down
when the controlling process dies and/or disconnects from the control
port. By using them we can avoid creating runaway Tor processes that
have to be killed with hacks.

Fixed a bug in SimplexMessagingIntegrationTest that should've caught
this.

This ensures that if two DatabaseExecutor tasks update the database and
broadcast events, the events are broadcast in the same order as the
updates occurred.

Patched jtorctl to pass more information about circuit events to the
controller to help us debug connection problems.

This may have been the cause of bug #69. Removed the wait-for-shutdown
code in DroidtoothPlugin that was added to attempt to avoid that bug.