Minor improvements in vertex to area linking

[vesameskanen]

Summary

• No through traffic properties are now assigned more accurately to created edges
• Area safety properties are now assigned as well
• Walkable area builder uses consistent parent entity in vertex generation. previously, a random entity from the area group was used.
• Some method renaming

Unit tests

One unit test added to verify that area safety factors are considered

NOTE: The test is a bit flaky, apparently because of double to float conversion. Test values are chosen to avoid value drifting.

Documentation

Javadoc added to explain the no through logic.

[leonardehrenfried]

Generally, I'm fine with this but would probably use a float everywhere.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 72.00%. Comparing base (c5fd182) to head (7864bd1).
⚠️ Report is 28 commits behind head on dev-2.x.

Additional details and impacted files

@@            Coverage Diff             @@
##             dev-2.x    #6826   +/-   ##
==========================================
  Coverage      71.99%   72.00%           
+ Complexity     19437    19426   -11     
==========================================
  Files           2100     2099    -1     
  Lines          78774    78729   -45     
  Branches        7963     7961    -2     
==========================================
- Hits           56717    56686   -31     
+ Misses         19249    19235   -14     
  Partials        2808     2808           

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

🚀 New features to boost your workflow:

• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[leonardehrenfried]

There are probably a few more places where we could use floats instead of doubles when constructing the graph but that is beyond the scope of the PR.

[vesameskanen]

I added one more somewhat related fix: walkable area builder passes the correct parent reference of each node when generating vertices. Previously part of the code used a common, randomly picked area group member. This did not have any effect because area group consists of entities at the same level. However, PR #6826 will introduce barrier logic which will get confused when node parent references are not consistent.

[miklcct]

It looks good now but give me a day or two to put this on our test bed to see if it still works well with the barrier PR.