Bug fix: ignore 0 in TDaTing cell match

[ritvje]

This PR fixes a bug in the pysteps.tracking.tdating.match function. As far as I can tell, in this bug the label 0 was included in possible match IDs, even though 0 denotes pixels that are not part of any cell. So there could be cases, where the advected cell would have e.g. 55% of 0 and 45% of some cell and it would not be matched to that cell. So this fix will first check if the advected cell contains only 0-labeled pixels, it is initiated as a new track. If not, then 0 is removed from the possible ID labels and the matching is done as previously.

[codecov[bot]]

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Project coverage is 83.33%. Comparing base (a471d49) to head (f84da89). Report is 1 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #350   +/-   ##
=======================================
  Coverage   83.32%   83.33%           
=======================================
  Files         161      161           
  Lines       12363    12368    +5     
=======================================
+ Hits        10302    10307    +5     
  Misses       2061     2061           

Flag	Coverage Δ
unit_tests	83.33% <100.00%> (+<0.01%)	:arrow_up:

Flags with carried forward coverage won't be shown. Click here to find out more.

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

[dnerini]

sorry for taking so long, I think we can merge this bugfix if @feldmann-m has no objections to this?