interpolating depth values

[Dirk Hohndel]

Hi Linus,

We have an interesting issue that I think may have been introduced by
commit 2b1017c986bf ("Interpolate depth for samples that have no depth").
It appears that this ends up throwing away data from dive computers that
for whatever reason recorded negative depth value.

Sure, negative depths seem weird and wrong, but we have at least one user
who is unhappy with the loss of information from the dive computer. And he
makes a somewhat compelling argument:
https://github.com/Subsurface-divelog/subsurface/issues/561

The part that I have to kinda agree with is that we used to keep those
values, and now we throw them away. And we seem to throw them away because
you decided that depth.mm == -1 is magical for the Suunto EON Steel. And
that seems like we are throwing away a lot of (possibly odd) data, but
data that we got from a dive computer, just to deal with an oddity of the
samples coming form the EON Steel.

Should we turn that magic value into something like -99999 and treat all
other negative values as just that - negative depth values?

/D