Fix unsigned Size type issues and migrate to C++20 standard

[chenpeizhi]

• Fix find_last() to return SIZE_MAX instead of -1 for unsigned Size type

• Add explicit casts for builtin function returns and fix sign-conversion warnings

• Migrate from GCC builtins to C++20 standard library bit operations

[Copilot]

Pull Request Overview

This PR modernizes the fbitset library by replacing GCC-specific compiler intrinsics with portable C++20 standard library bit manipulation functions from the <bit> header.

Key Changes:

• Replaced __builtin_clz, __builtin_ctz, and __builtin_popcount intrinsics with std::countl_zero, std::countr_zero, and std::popcount respectively
• Changed find_last() return value from -1 to std::numeric_limits<Size>::max() for consistency with unsigned type semantics
• Updated build requirements to C++20 and CMake 3.12

Reviewed Changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 1 comment.

File	Description
include/fbitset.hpp	Replaced GCC intrinsics with C++20 <bit> header functions; updated find_last() return value and documentation
test/basic.cpp	Added explicit Size() casts in initializer list; updated test assertion for new find_last() return value
README.md	Updated documentation to reflect use of C++20 <bit> header instead of GCC intrinsics
CMakeLists.txt	Updated minimum CMake version to 3.12, C++ standard to 20, and added CMAKE_CXX_STANDARD_REQUIRED

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.