Peter M. Groen / FlashMQ

Empty leaves weren't removed, causing increase in memory use.

This method incurs no extra CPU load when messages aren't dropped.

One was confirmed: writing an mqtt packet into a client that
disconnected after checking the weak pointer for validity.

The rest made sense to change as well.

I have no explanation for it, but it happens (with QMQTT, should it
matter).

It was a mistake that it ran inside the timer thread, which came to
light by asserting the thread local authentication pointer (see a few
commits ago).

There were bugs in which authentication object was used when, causing
threadings bugs. Instead of getting from the 'sender', we can just store
a thread local pointer.

Files are simple serialized bytes prefaced by lengths. File is hashed to
verify integrity. This was also a good way preventing unexpected errors
when trying to crash the parser by having it load a different file.

This change includes some refactoring that was necessary:

- It 'fixes' looking at the wrong thread's authentiction. This is still
  wrong though. It will be fixed by a thread local pointer in the next
  commit.
- Deadlocks with yourself are handled in rwlockguard.
- QoSPacketQueue is now a class.
- Probably other tweaks.

It didn't clean up the old ones, causing unpredictable problems.

This fixes side effects when system time changes.

This performs a whole lot better.

Three parts to it:

- Start the app fresh per test. This avoids annoyances like getting
  retained messages on subscribe, messing up the test.
- Waiting for suback was apparently necessary.
- Because the Qt event loop was given time, waiting for publishes was
  sometimes pointless because it had already arrived. So, checking the
  receive list first.

Fix version bug.

Also include a few stats.