dive site handling update

[Dirk Hohndel]

Just in case people are wondering what I'm doing...

I'm about thirty-some-odd commits and two nights with way too little sleep
into rewriting our complete location / coordinates / dive site handling
code.

This involves everything from the data structures used, addition of a ton
of helper functions, minor UI changes (the coordinates disappear from the
MainTab - they were only there to be able to edit them manually for a
dive, but this will now move to the dive sites UI), changes to the logic
of a ton of operations, to modifications of the storage formats.

There are still functionality gaps (right now it looks like you can edit
dive sites, but actually you can't; I'm still fighting to make the
Subsurface Webservice GPS fix import do the right thing again) and bugs (I
have at least one XML file that is not parsed correctly, a minute ago I
fixed a hard crash, etc). But it's getting closer.

When this lands in master it will make irreversible changes to your XML
files (people using git storage will smile as they can always revert
back). I actually am so worried about this that I revved our git and XML
file format version to 3 and added a special backup function that creates
an .xml-v2 (or .srff-v2) backup file if you had opened an XMLv2 file and
subsequently save this as v3 file. But even with that I want people to be
keenly aware that this almost certainly contains a bunch of bugs and that
it's possible that it will damage your data file.

I have spent quite a bit of time on testing things and constantly find
and fix issues - and I bet there are many more.

This is at least another day or two away from being ready for master, so
the email is just a "fair warning". I'll post another warning when I
actually push this - and I will not create "regular daily builds" for a
few days until more people had a chance to test the code.

Having said all this, I'm quite excited. I think this will be a very cool
new feature. There are lots of UI and work flow issues that still need to
be solved, but I think it's going in a good direction.

/D

PS: no, I haven't forgotten 4.4.1. We just haven't seen as much use of 4.4
as I would have liked to see in order to make sure we have shaken out
enough of the bugs that might still be hiding - based on my estimates
(which might be off) only about 10-15% of our users have updated so far.

Given how much of my time has been absorbed by the dive site changes
(basically all of my Subsurface time in the past three days, errr, nights)
I just figured I might as well delay 4.4.1 until next week.