Better type return value of json.detect_encoding

[jonathandung]

It would be beneficial to annotate the return type of json.detect_encoding as the values that are actually possible, so that the user knows approximately what the function does just by looking at the stub. This would mean changing the signature from

def detect_encoding(b: bytes | bytearray) -> str: ...  # undocumented

to

def detect_encoding(b: bytes | bytearray) -> Literal['utf-8', 'utf-8-sig', 'utf-16', 'utf-16-be', 'utf-16-le', 'utf-32', 'utf-32-be', 'utf-32-le']: ...  # undocumented

.

See the relevant source.

[github-actions]

According to mypy_primer, this change has no effect on the checked open source code. 🤖🎉