Add rotation on constraint solving

[fallenatlas]

Description

Adds angular velocity computation in penetration constraint solving.

Checklist

• Self-review changes.
• Evaluate impact on the documentation.
• Add entry to the changelog's unreleased section.

[github-actions]

PR Preview Action v1.4.8
🚀 Deployed preview to https://GameDevTecnico.github.io/cubos/preview/pr-1315/
on branch gh-pages at 2024-10-07 19:42 UTC

[github-actions]

⚠️ Clang-Tidy found issue(s) with the introduced code (1/1)

[codecov]

Codecov Report

Attention: Patch coverage is 2.80899% with 173 lines in your changes missing coverage. Please review.

Project coverage is 40.73%. Comparing base (ada20d7) to head (22dd9a0).
Report is 7 commits behind head on main.

Files with missing lines	Patch %	Lines
...c/physics/solver/penetration_constraint/plugin.cpp	0.00%	134 Missing ⚠️
engine/src/collisions/narrow_phase/plugin.cpp	4.54%	21 Missing ⚠️
...src/physics/constraints/penetration_constraint.cpp	0.00%	16 Missing ⚠️
engine/src/physics/plugin.cpp	0.00%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1315      +/-   ##
==========================================
- Coverage   40.84%   40.73%   -0.12%     
==========================================
  Files         430      430              
  Lines       30409    30500      +91     
==========================================
+ Hits        12421    12423       +2     
- Misses      17988    18077      +89     

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

[RiscadoA]

Thanks for working on this 🥳 Very cool to see rotation being relevant. As usual, I don't really understand what's going on here, so I'll just trust you on the math.
The sample also runs fine on my machine, but I noticed the movement of the boxes is sometimes weird/unnatural? Is it still missing step or something? What's next?

[RiscadoA]

Also, don't forget to add an entry to the changelog!

[fallenatlas]

Thanks for working on this 🥳 Very cool to see rotation being relevant. As usual, I don't really understand what's going on here, so I'll just trust you on the math. The sample also runs fine on my machine, but I noticed the movement of the boxes is sometimes weird/unnatural? Is it still missing step or something? What's next?

Sometimes it still looks weird yeah, I'm not entirely sure of what the problem is, it could be because of the friction and bounciness settings in the sample which could produce something we don't naturally expect, or my guess is that it has to do with the softness parameters, which might be too strong (they limit the velocities for the solving thus making it feel less crispy and more sluggish, but it's important for stability), but I don't fully understand this part yet so it's something that I'll have to look into in the future.

[RiscadoA]

@fallenatlas Just noticed your commit messages are not very correct:

• when working on samples, don't use feat(samples), instead use something like docs(physics)
• don't use penetration_constraint as a scope, use physics (the rule is to stick to the B-label names). If you want to mention it's related to that, put it in the message itself.

If you can, change this before merging!