This is needed for Android since it needs to communicate with java apis
which require everything on the main thread. The additional thread was
not really needed since you can implement using a signal based approach
which ends up being the same thing.

Still not working right...

Link ids were being used interchangeably with the mavlink channel for
the link. Link ids handed out as a numerically increasing integer. Once
the counter went past MAVLINK_COMM_NUM_BUFFERS all hell would break
loose and cause memory corruption.

If it doesn’t exist it will be prefilled with all registered categories
set to false

Changed settings version as the link configurations must be redone (with the new port names)

Also converted qDebug statements to logging category

The goal of this first installment is to organize the code a bit so it's more readable,
clean up a bit of left over cruft, and manage link configurations (and links in general).

There's also a partial fix for the databits and stopbits. I'm unsure how to deal with the logic
around the 'is_cdc' variable, as I can't really follow what it's for.

Moving methods from MainWindow to QGCApplication so that more things
can run without a window. Helpful for testing.

Plus a large number of other changes to allow for orderly shutdown of
objects without crashes or asserts. This is need for unit tests to
create/delete global state around every test.

This function is called automatically when a QSettings object is destroyed,
so is rarely needed to be called explicitly.

There are still some cases where
a QSettings object is passed-by-reference, so there I left the sync() calls as
it's unclear how permanent those objects are and if they'd call sync() before
application shutdown

Centralize all connecting and disconnecting of links through
LinkManager::[Connect|Disconnect]Link api. Connect/Disconnect no longer
allowed through LinkInterface. This allows creation of new
LinkManager::setConnections[Suspended|Allowed] methods to disallow
connections as needed.

This api doesn’t make any sense given the fact that LinkInterface
signals bytesReceived with the bytes in the signal. Also the reason why
it is never called!

This reverts commit b4089ad9.

Replaced with only 1 output message with the result of the first serial port scan.

Fix execution context of thread routines - note that this is not quite the architecture the Qt folks intended, but as they admit its the consequence of their architecture.

Avoid evil port config for CDC devices altogether, non-CDC devices (unknown devices) should be unaffected

Do not configure the port if talking to PX4 units (since USB does not know anything about baud rate)

Also commented out noisy serial link error output