Update the benchmark to load datasets from multiple S3 buckets when None is specified

[R-Palazzo]

Resolve #604
CU-86b9zwpcu

This PR includes:

• Relying on SDV download_demo() to get sdv_dataset. This works for public datasets or for users with SDV-Enterprise installed. Otherwise, because on SDV there is some validation to allow access to the public bucket only, I reused the SDV logic here, so it works if the S3 client is created with valid credentials (from input keys or env variables).
• Because we rely on download_demo(), we no longer save the dataset locally.
• Improvement in the _generate_job_arg_list: Now we're downloading datasets only once instead of once per job before.
• Added an integration test with a private dataset. Also tested it on AWS here and GCP here.

[sdv-team]

Task linked: CU-86b9zwpcu SDGym - Update the benchmark to load datasets from multiple S3 buckets when None is specified #604

[codecov]

Codecov Report

❌ Patch coverage is 93.24324% with 5 lines in your changes missing coverage. Please review.
✅ Project coverage is 85.80%. Comparing base (c90bbce) to head (261893a).

Files with missing lines	Patch %	Lines
sdgym/datasets.py	92.30%	4 Missing ⚠️
sdgym/benchmark.py	95.45%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #607      +/-   ##
==========================================
+ Coverage   85.67%   85.80%   +0.13%     
==========================================
  Files          40       40              
  Lines        3679     3742      +63     
==========================================
+ Hits         3152     3211      +59     
- Misses        527      531       +4     

Flag	Coverage Δ
integration	43.85% <81.08%> (+0.25%)	⬆️
unit	81.32% <78.37%> (-0.17%)	⬇️

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

☔ View full report in Codecov by Sentry.
📢 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.