Meta issue: improve documentation for implementation of the PEP 3118 in Python

Documentation

Recent d.p.o discussion thread and SC decision reveal some problems with current docs. Affected parts are the memoryview and struct/array/ctypes modules. I'll list issues in a random order with a hope that people will extend/correct this list:

• The struct module docs has Format Characters section. Though, some type codes now have more than one character. We should rename this section and use some different term. Format Codes? // https://github.com/python/cpython/pull/149483
• array docs says: The type is specified at object creation time by using a type code, which is a single character. // https://github.com/python/cpython/pull/149483
• memoryview docs are too vague in their references to the struct module:
  • "format" described as being "in struct module style"
  • it might be not clear that means "single element native format" (C type with native byte size, native endianness, and native alignment)
• ctypes has no documentation for the buffer protocol support at all; though, it's probably most advanced across of the stdlib, see Lib/test/test_ctypes/test_pep3118.py.
• ctypes._SimpleCData._type_ docs says "Class attribute that contains an internal type code, as a single-character string. See Fundamental data types for a summary." // https://github.com/python/cpython/pull/149483

Linked PRs

• gh-149483

Empieza por las secciones de documentación de memoryview, struct, array y ctypes; después, lee la discusión enlazada y la decisión del SC sobre la terminología prevista de PEP 3118. Usa Lib/test/test_ctypes/test_pep3118.py para comprender el comportamiento existente de ctypes con el buffer protocol. La tarea está terminada cuando las referencias a los formatos de memoryview sean precisas y la compatibilidad de ctypes con buffers esté documentada.