Add min/max/clamp/signum helpers for Float/Double/Int64/UInt

[bobzhang]

Motivation

• Numeric types lacked a consistent set of small utility methods (min, max, clamp, signum) across Float, Double, Int64, and UInt, forcing callers to hand-roll common helpers.
• Provide IEEE-style min/max semantics for floating types (when exactly one operand is NaN, return the non-NaN operand) and parity with existing Int APIs to simplify client code.

Description

• Added Float::min, Float::max, and Float::signum to the float package implementing IEEE-style NaN handling and parity with Double::signum (float/methods.mbt).
• Added Double::min and Double::max with the same NaN semantics (builtin/double.mbt).
• Added Int64::min, Int64::max, and Int64::clamp (builtin/int64.mbt) and UInt::min, UInt::max, and UInt::clamp (builtin/uint.mbt).
• Added tests covering basic behavior, NaN cases for float/double min/max, and clamp panic cases for invalid ranges; updated generated interface files (.mbti) to expose the new APIs.

Testing

• Ran moon fmt which completed successfully.
• Ran moon info and moon check which completed successfully and updated package interfaces.
• Ran the full test suite with moon test and observed all tests passed (Total tests: 5790, passed: 5790, failed: 0).

---

Codex Task

---

[devin-ai-integration]

✅ Devin Review: No Issues Found

Devin Review analyzed this PR and found no potential bugs to report.

View in Devin Review to see 4 additional findings.

[coveralls]

Pull Request Test Coverage Report for Build 2737

Details

• 53 of 57 (92.98%) changed or added relevant lines in 4 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage decreased (-0.01%) to 95.668%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
builtin/int64.mbt	10	12	83.33%
builtin/uint.mbt	10	12	83.33%

Totals
Change from base Build 2735:	-0.01%
Covered Lines:	12985
Relevant Lines:	13573

---

💛 - Coveralls