fix TripTime NullPointerException

[miklcct]

Summary

Fixes NullPointerException on TripTime. The value is clearly nullable but no null check took place, causing NullPointerException.

Issue

Fixes #6088

Unit tests

No idea yet.

Documentation

None needed

Changelog

Bumping the serialization version id

[miklcct]

I have tested that this does prevent error coming from the API, and my frontend looks good. However, my frontend only uses the content for some logic and it didn't show the contents of affected trips.

[codecov]

Codecov Report

Attention: Patch coverage is 0% with 4 lines in your changes missing coverage. Please review.

Please upload report for BASE (dev-2.x@155d19b). Learn more about missing BASE report.
Report is 160 commits behind head on dev-2.x.

Files with missing lines	Patch %	Lines
...java/org/opentripplanner/model/TripTimeOnDate.java	0.00%	3 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             dev-2.x    #6094   +/-   ##
==========================================
  Coverage           ?   69.84%           
  Complexity         ?    17454           
==========================================
  Files              ?     1975           
  Lines              ?    74604           
  Branches           ?     7636           
==========================================
  Hits               ?    52107           
  Misses             ?    19846           
  Partials           ?     2651           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[optionsome]

We should probably get an idea what is the original data(?) problem that causes this in the first place. There might be better way to solve this specific problem.

[vpaturet]

The root cause of the issue seems to be a bug in the GraphQL API where the scheduled timetable is used to resolve a real-time updated trip, leading to data inconsistency (to be confirmed and further analyzed, see org.opentripplanner.apis.gtfs.datafetchers.TripImpl.stopTimes())

We discussed the fix proposed in this PR during the developer meeting and agreed that handling the null case would hide a potential bug and propagate inconsistent data down the stack.

In this particular case we suggest addressing the root cause of the bug rather than handling the NPE.

[t2gran]

We will close this next week unless we get more insight/feedback on this.