Fix/schemes again

[J08nY]

This significantly improves the CC scheme extraction by:

• Fixing the extraction of several schemes that were mixing
  certified and archived entries by accident.
• Improving the extraction of cert_ids from scheme sites.
• Improving the matching heuristic to consider more attributes
  that are usually present in the site data.

Also adds an evaluation notebook to see how this performs.

Fixes #454
Fixes #455

[codecov]

Codecov Report

Attention: Patch coverage is 77.77778% with 102 lines in your changes missing coverage. Please review.

Project coverage is 67.89%. Comparing base (f2c654f) to head (c5ceab2).
Report is 20 commits behind head on main.

Files with missing lines	Patch %	Lines
src/sec_certs/sample/cc_scheme.py	77.78%	76 Missing ⚠️
src/sec_certs/model/cc_matching.py	73.44%	17 Missing ⚠️
src/sec_certs/sample/cc_certificate_id.py	42.86%	8 Missing ⚠️
src/sec_certs/dataset/cc.py	50.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #456      +/-   ##
==========================================
+ Coverage   67.52%   67.89%   +0.38%     
==========================================
  Files          62       62              
  Lines        7609     7900     +291     
==========================================
+ Hits         5137     5363     +226     
- Misses       2472     2537      +65     

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