Polish benchmarks page

[philippmwirth]

Polish benchmarks page

Clean up the benchmarks page in the docs:

• Use consistent names (e.g. CIFAR-10 and kNN).
• Fix names (e.g. Imagenet100 -> ImageNet100).
• Use consistent structure: Dataset and code on top, then a short description, and then the results table.
• Order by relevancy (ImageNet, ImageNet100, Imagenette, CIFAR-10).
• Point to the examples and tutorials in the next steps section.
• Make README consistent with benchmarks page.

To review it's best to build the docs locally and go through the page:

git pull philipp-lig-3098-cleanup-pip-benchmark-docs
cd docs
make html-noplot
python -m http.server 1234

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (514ffd7) 85.24% compared to head (371da53) 85.24%.
Report is 1 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #1432   +/-   ##
=======================================
  Coverage   85.24%   85.24%           
=======================================
  Files         130      130           
  Lines        5517     5517           
=======================================
  Hits         4703     4703           
  Misses        814      814           

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

[guarin]

Thanks LGTM! Left some comments but feel free to resolve and merge!