Take realtime patterns into account when storing realtime vehicles

[leonardehrenfried]

Summary

When storing realtime vehicle positions by pattern the realtime added patterns are not taken into account correctly.

This PR uses the original pattern as the key for the storage map.

Unit tests

Added.

Documentation

Explained it in Javadoc.

[codecov]

Codecov Report

Attention: Patch coverage is 50.00000% with 2 lines in your changes missing coverage. Please review.

Project coverage is 69.73%. Comparing base (fbca53d) to head (e81aa53).
Report is 65 commits behind head on dev-2.x.

Files	Patch %	Lines
...hicles/internal/DefaultRealtimeVehicleService.java	50.00%	1 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             dev-2.x    #5994   +/-   ##
==========================================
  Coverage      69.73%   69.73%           
- Complexity     17294    17297    +3     
==========================================
  Files           1954     1954           
  Lines          74163    74167    +4     
  Branches        7595     7595           
==========================================
+ Hits           51715    51719    +4     
  Misses         19809    19809           
  Partials        2639     2639           

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

[optionsome]

There are potential issues with using the originalTripPattern from a realtime added trip pattern but I think so far we have kinda merged the realtime added patterns with the normal ones in API if I remember correctly so this does make sense. I think in the somewhat near future we do want to refactor the patterns a bit more so there is a pattern which makes more sense for clients to use in most cases.