The DataThread shouldn't have a MML all the time, but it does need it
when signaling that the data source is not found.

Also fixed a line in the FilterViewport that was causing the enabled
status of editors to fail to update.

The reason between the UI interactions with the DataThread was simple: it
was using a type of lock that's meant for interacting with the UI.

This seemed to fix all the issues with dropped samples, but we'll still
have to run it through more rigorous tests in the future.

Changes to the DataThread, SourceNode, and UIComponent classes allow data
threads to disable callbacks if they are no longer receiving input. So
far this has only been tested with the Intan Board, but it works really well.
Unplugging the board in the middle of acquisition allows the thread to exit,
and callbacks to be disabled, without any seg faults.

DataThreads (e.g. IntanThread, FileReaderThread, FPGAThread) are now created at
the same time as the source node, rather than at the start of data acquisition.
New methods for starting/stopping individual threads are required, although
only the appropriate methods for the IntanThread have been written.

Another important change is that the SourceNode now periodically checks for
an appropriate input source every few seconds while acquisition is not in
progress. It's the responsibility of the individual DataThreads to notify
the SourceNode if their input has disappears. In the case of the IntanThread,
this involves attempting to change the baud rate. If an error code returns,
it informs the SourceNode that the input is missing. This, in turn, informs
the FilterViewport that the source is no longer enabled, thus deactivating
that particular signal chain.