Correctly parse multiple tags in the divelog

The old parsing code overwrote the first comma with a '\0' and then
checked the index against the length of the buffer - which was changed by
replacing the ',' with the '\0'.

This means that since commit 78acf20848c0 ("Don't crash on loading tags
longer than 127 chars") Subsurface has potentially damaged / lost data in
dive files!

Added a test dive that shows the issue if opened by a Subsurface version
after the commit mentioned above but before this commit.

Reported-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>

0 files changed, 0 insertions, 0 deletions