Thread related crash on failed dive computer download

[Berthold Stoeger]

On Dienstag, 31. Oktober 2017 21:00:37 CET Berthold Stoeger wrote:
> So in the attachment is a different approach: don't output errors if not in
> the main thread. Someone who generates a thread is responsible of flushing
> the errors later. Sounds much less brittle to me.

It works in my limited test-case and I generated a pull request:
https://github.com/Subsurface-divelog/subsurface/pull/753

Fixing the race condition should not be too hard.

Berthold