Update Statsmodels documentation (0.15.0)

[ridge-kimani]

Updates Statsmodels from 0.9.0 to 0.15.0 (released August 27, 2026) and moves the URLs to https.

The site now uses pydata-sphinx-theme, so the filters needed updating for the new markup:

• the content container is the article; the old .body no longer exists
• the type of API pages comes from the breadcrumbs ("User Guide > chapter > ...") instead of the old .related bar, which gives the same chapter names as before (plus the chapters added since 0.9, such as Vector Autoregressions and Generalized Additive Models)
• notebook example pages lose their In [1]: prompts and cell wrappers, and output cells become plain <pre> blocks
• [source] links (to module pages that aren't scraped) are dropped, and parameter names are separated from their type classifier (endog: array_like instead of endogarray_like)
• py-modindex.html is skipped, and the attribution years follow LICENSE.txt

Regenerating also picks up the MathFilter from #2752 (tracked in #2763).

Verified locally with thor docs:generate statsmodels --force:

• 6,989 pages generated without errors (the 0.9.0 docs had about 3,500); all 26 types of the production index are still there, with 32 types and 6,988 entries in total
• 576 pages contain math, 2,791 <math> elements, 3 fallbacks that are errors in the upstream TeX (\endog_i in two loglike_obs pages, and a notebook redefining \r with \newcommand), which MathJax rejects as well
• Checked user guide chapters (regression, glm), API pages (generated/statsmodels.genmod.families.family.Gamma.loglike_obs) and notebook examples (examples/notebooks/generated/ols) in the app

If you're updating existing documentation to its latest version, please ensure that you have:

• Updated the versions and releases in the scraper file
• Ensured the license is up-to-date (still BSD-3-Clause; copyright years updated from LICENSE.txt)
• Ensured the icons and the SOURCE file in public/icons/your_scraper_name/ are up-to-date if the documentation has a custom icon
• Ensured self.links contains up-to-date urls if self.links is defined
• Tested the changes locally to ensure:
  • The scraper still works without errors
  • The scraped documentation still looks consistent with the rest of DevDocs
  • The categorization of entries is still good

🤖 Generated with Claude Code