chore(deps): update dependencies

[TestingPlant]

No description provided.

[github-actions]

Benchmark Results for general

ray_intersection/aabb_size_0.1                     [  20.8 ns ...  20.8 ns ]      -0.09%
ray_intersection/aabb_size_1                       [  20.6 ns ...  20.8 ns ]      +0.89%
ray_intersection/aabb_size_10                      [  20.5 ns ...  20.5 ns ]      +0.16%
ray_intersection/ray_distance_1                    [   4.2 ns ...   4.2 ns ]      +0.13%
ray_intersection/ray_distance_5                    [   4.2 ns ...   4.2 ns ]      -0.18%
ray_intersection/ray_distance_20                   [   4.2 ns ...   4.2 ns ]      -0.19%
overlap/no_overlap                                 [  28.6 ns ...  28.7 ns ]      +0.44%
overlap/partial_overlap                            [  28.6 ns ...  28.6 ns ]      -0.13%
overlap/full_containment                           [  21.7 ns ...  21.7 ns ]      -0.08%
point_containment/inside                           [   4.9 ns ...   4.9 ns ]      -0.12%
point_containment/outside                          [   4.4 ns ...   4.4 ns ]      +0.13%
point_containment/boundary                         [   5.0 ns ...   5.0 ns ]      +0.01%

Comparing to 830c5fd

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 22.24%. Comparing base (830c5fd) to head (00ad881).
⚠️ Report is 11 commits behind head on main.

@@            Coverage Diff             @@
##             main     #926      +/-   ##
==========================================
+ Coverage   22.20%   22.24%   +0.03%     
==========================================
  Files         161      161              
  Lines       15992    16234     +242     
  Branches      488      498      +10     
==========================================
+ Hits         3551     3611      +60     
- Misses      12343    12521     +178     
- Partials       98      102       +4     

Files with missing lines	Coverage Δ
crates/hyperion-proxy/src/lib.rs	0.00% <ø> (ø)

... and 6 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[github-actions]

Benchmark Results for general

ray_intersection/aabb_size_0.1                     [  20.5 ns ...  20.6 ns ]      +0.45%
ray_intersection/aabb_size_1                       [  20.5 ns ...  20.6 ns ]      +0.50%
ray_intersection/aabb_size_10                      [  20.5 ns ...  20.5 ns ]      +0.41%
ray_intersection/ray_distance_1                    [   4.2 ns ...   4.2 ns ]      -0.14%
ray_intersection/ray_distance_5                    [   4.2 ns ...   4.2 ns ]      -0.05%
ray_intersection/ray_distance_20                   [   4.2 ns ...   4.2 ns ]      -0.29%
overlap/no_overlap                                 [  28.6 ns ...  28.7 ns ]      +0.49%
overlap/partial_overlap                            [  28.6 ns ...  28.7 ns ]      +0.19%
overlap/full_containment                           [  21.6 ns ...  21.6 ns ]      -0.02%
point_containment/inside                           [   4.9 ns ...   4.9 ns ]      -0.21%
point_containment/outside                          [   4.4 ns ...   4.4 ns ]      -0.06%
point_containment/boundary                         [   4.9 ns ...   4.9 ns ]      -0.10%

Comparing to 830c5fd

[andrewgazelka]

Closing this, and it is not a judgement on the change.

The project has moved a long way in a short time and is being built quite
differently now. The internals this touches have shifted underneath it, so
rather than leave the pull request sitting open against a tree it no longer
matches, I would rather close it honestly and say why.

Some of what has changed, so you can judge whether it is worth revisiting:

• The engine has been reorganised around a strict separation between modules
  that register components and modules that install behaviour. CLAUDE.md in
  the repository root documents the convention and why it exists.
• The smash game mode has had substantial work: every ability now draws a
  visual effect on the press, and several tests that could not fail have been
  replaced with ones that were watched failing before being trusted.
• Continuous integration no longer runs automatically. Every check is still
  there and nix run .#ci runs all of them locally, which was always the same
  command CI ran. The workflow files explain the reasoning.

If you still want the change in, a fresh pull request against current main
is the cleanest path, and running nix run .#ci locally first will tell you
what a reviewer would have seen. Genuinely, thank you for taking the time on
this one.

(sent by an AI agent via Claude Code, on behalf of the maintainer)