[rudycelekli]

Summary

Forward the modulus argument from Float.__pow__ and Float.__rpow__ to Python's native float methods.

Currently pow(item(2.0), 3, 3) returns 8.0, silently ignoring the modulus. The equivalent native-float call raises TypeError, because modular exponentiation is supported only for integers. Forwarding the argument preserves that native error rather than returning an unrelated numeric result. Two-argument powers and an explicit None modulus retain their existing behavior.

Duplicate Check

• I searched the open PRs and found none that already fixes this issue.
• Related open PR: Preserve floating-point results of negative integer powers #640 fixes Integer power result types. This change is limited to forwarding arguments in the separate Float methods and does not modify Integer powers.

Validation

• Native float methods provide the expected-result oracle.
• Unchanged implementation: six new invalid-modulus cases fail; two None-modulus controls pass.
• Corrected implementation: all 120 item tests pass; the complete suite passes 1,232 tests on Python 3.13.5, with the pinned toml-test submodule checked out.
• Ruff check, Ruff format check, and git diff --check pass for the changed files.

Agent Drafting Metadata

• Agent: OpenAI Codex
• Model: inherited Codex subscription model; exact model identifier was not exposed to this worker.
• Notes: AI-assisted implementation and tests. A separate agent inspected the diff and independently ran all eight power regressions. No human review or maintainer acceptance is claimed.