[djbower]

Summary

Documentation updates for v2, plus one example fix that follows from #163.

• Guides:
  • Add a nested solvers section, and update the troubleshooting page and common tasks.
  • Update the installation and developer guides for v2.
  • Point questions and ideas to GitHub Discussions.
• Output page: correct the IW shift and constraint descriptions.
• References:
  • Add Polman et al. (2026) to the list of works that use Atmodeller.
  • Match the FastChem 2 title to the published version.
• Style: British spelling in prose.
• Condensate data ranges (follow-up to Condensates are not stable outside the temperature range of their data #163):
  • troubleshooting.rst and common_tasks.rst ("Add thermodynamic data") now say that gas data are extrapolated outside their temperature range, whereas condensates are not stable there by default. They also explain how to opt back in with extrapolate_below/extrapolate_above.
• Atmosphere example fix:
  • When the condensed data were split into solid and liquid states, the example's O2Si_cd and Si_cd became O2Si_s and Si_s, which dropped the liquid field. Its 3000 K surface is above the melting points of silica (1996 K) and silicon (1690 K).
  • The example now includes both phases, and each can only form within the temperature range of its data.
  • Liquid silica and silicon are now stable at the surface, with the solids at the cooler levels. Before this, no condensate was stable at 3000 K.
  • An audit found no other use of the old combined state that lost a liquid phase.

Testing

• The documentation builds as in the docs CI job (sphinx-apidoc, then sphinx-build -W --keep-going) with no warnings. This includes executing all the example notebooks.
• The atmosphere example solves all four levels on the first attempt.

🤖 Generated with Claude Code