Use size_t for sizes and indices instead of ptrdiff_t

[mizdebsk]

Replace uses of std::ptrdiff_t with std::size_t for variables and parameters representing sizes and positions in string and span-like containers. This better reflects intent, avoids unnecessary signed/unsigned conversions, and aligns with common C++ conventions.

[mkoncek]

I simply follow guidelines and a simple rule: use unsigned only when you want modulo arithmetic on overflow.

• https://github.com/isocpp/CppCoreGuidelines/blob/master/CppCoreGuidelines.md#es106-dont-try-to-avoid-negative-values-by-using-unsigned
• https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2019/p1428r0.pdf
• https://youtu.be/Puio5dly9N8?t=733

I adopted this approach in most of my projects.

[mizdebsk]

I still think that size_t (or ssize_t) is the natural way of expressing sizes, not ptrdiff_t.
But if you disagree feel free to close the PR.

[mkoncek]

I want to keep everything signed.
ssize_t is a POSIX type, ptrdiff_t is a standard type (returned by the ssize functions.