Update the old plots in the README.md and add benchmarks to docs

[efaulhaber]

This PR

• updates the old plots of runtime by implementation in the README.md. The update benchmark now also shows the update strategies of the grid NHS.
• adds the benchmark suite to the doc.
• sets sort_neighbor_lists = true in the benchmarks. This is the default, but having this kwarg allows us to trixi_include this benchmark and setting it to false until AMDGPU.jl properly implements sorting.

Preview at https://trixi-framework.org/PointNeighbors.jl/previews/PR170/.
Note that the README.md doesn't show the new plots until merged into main, but they can be previewed on the starting page of the docs linked above.

[codecov]

Codecov Report

❌ Patch coverage is 40.00000% with 3 lines in your changes missing coverage. Please review.
✅ Project coverage is 83.94%. Comparing base (10a389d) to head (4d16024).

Files with missing lines	Patch %	Lines
test/point_cloud.jl	40.00%	3 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #170      +/-   ##
==========================================
- Coverage   84.59%   83.94%   -0.66%     
==========================================
  Files          15       15              
  Lines         792      791       -1     
==========================================
- Hits          670      664       -6     
- Misses        122      127       +5     

Flag	Coverage Δ
unit	83.94% <40.00%> (-0.66%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

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

🚀 New features to boost your workflow:

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

[svchb]

Why are there no CPU results for TLSPH?

[efaulhaber]

Why are there no CPU results for TLSPH?

It's in the comments. I would like to first merge trixi-framework/TrixiParticles.jl#1256 for better performance on large CPUs. I will add these numbers in the next PR.

[efaulhaber]

Codex Review

Adds CPU benchmark data and plotting for WCSPH implementations and update strategies, generates those plots during documentation builds, and refreshes README benchmark content.

This is an AI-generated code review. Please verify the findings and summary before acting on them.
_{Review generated by codex-pr-review}

[efaulhaber]

Codex Review

The PR adds update-strategy and implementation benchmark runners, plots, documentation, and published benchmark assets. The new update comparison does not use identical workloads across strategies and can fail for supported radius values because of a fixed neighbor-list capacity.

This is an AI-generated code review. Please verify the findings and summary before acting on them.
_{Review generated by codex-pr-review}

[efaulhaber]

Codex Review

The PR expands benchmark runners and plots, adds update-strategy comparisons and benchmark documentation, and improves deterministic point-cloud perturbation.

This is an AI-generated code review. Please verify the findings and summary before acting on them.
_{Review generated by codex-pr-review}