Replace LIMIT_AS_RANGE macro with constexpr function

[konard]

📝 Summary

This PR replaces the LIMIT_AS_RANGE macro with a modern constexpr template function, addressing issue #76.

🔄 Changes Made

• Removed: LIMIT_AS_RANGE macro and its #undef directive
• Added: Internal::FullRange<T>() constexpr template function in the Internal namespace
• Updated: All predefined range constants (SByte, Int16, Int32, etc.) to use the new function

✨ Benefits

• Type Safety: Template function provides compile-time type checking vs. macro expansion
• Debugging: Function calls are easier to debug and step through than macro expansions
• Namespace Safety: Function is contained within Internal namespace, avoiding global pollution
• Modern C++: Follows modern C++20 best practices by using constexpr functions over macros
• Maintainability: Template function is more readable and maintainable than macro

🏗️ Implementation Details

The original macro:

#define LIMIT_AS_RANGE(type) std::numeric_limits<type>::lowest(), std::numeric_limits<type>::max()

Has been replaced with:

namespace Internal
{
    template<typename T>
    constexpr Range<T> FullRange() noexcept
    {
        return Range<T>(std::numeric_limits<T>::lowest(), std::numeric_limits<T>::max());
    }
}

Usage changed from:

constexpr auto SByte = Range(LIMIT_AS_RANGE(std::int8_t));

To:

constexpr auto SByte = Internal::FullRange<std::int8_t>();

✅ Testing

• ✅ Code compiles successfully with C++20 standard
• ✅ All predefined ranges (SByte, Int32, UInt64, etc.) produce correct min/max values
• ✅ Template function works correctly for all numeric types (signed/unsigned integers, floats)

📋 Fixes

Closes #76

---

🤖 Generated with Claude Code