Add alphabet-sort

[mikasenghaas]

Description

Moved from research-environments (PR)

Type of Change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update
• Test improvement

Testing

• All existing tests pass when running uv run pytest locally.
• New tests have been added to cover the changes

Checklist

• My code follows the style guidelines of this project as outlined in AGENTS.md
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• Any dependent changes have been merged and published

Additional Notes

---

Note

Adds a new multi-turn evaluation environment for alphabetically sorting names with dataset-backed prompts and power-scaled similarity scoring.

• New alphabet-sort environment builds HF-driven multi-turn sessions, randomizes first/last-name sorting, and tags newly added names per turn
• Implements XML-tagged IO (alphabetical_sorted then combined_alphabetical_sorted) and multi-attempt scoring that requires monotonic improvement
• Reward uses difflib similarity with configurable exponent and power_per_turn option; exposed via vf.MultiTurnEnv rubric
• Includes README with quickstart/args and pyproject.toml (v0.1.10, verifiers>=0.1.8)

^{Written by Cursor Bugbot for commit 8054316. This will update automatically on new commits. Configure here.}