Stdlib: add 'obvious' default values

[AlexWaygood]

All parameters annotated with Literal[False] must, logically speaking, have a default of False (if they have a default value at all). The same goes for parameters annotated with None, Literal[True], Literal[0], Literal["foo"], etc. etc.

I ran this code in the interactive REPL to create this PR:

>>> from pathlib import Path
>>> import re
>>> for path in Path("stdlib").rglob("*.pyi"):
...     source = path.read_text()
...     new_source = re.sub(r": Literal\[([a-zA-Z0-9_\"]+)\] = \.\.\.", r": Literal[\1] = \1", source)
...     new_source = re.sub(r": None = \.\.\.", ": None = None", new_source)
...     if new_source != source:
...         path.write_text(new_source)

[github-actions]

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

[srittau]

Didn't check against the implementation, but the reasoning (and spot checking) is sound.

[AlexWaygood]

I ran this code in the interactive REPL to create this PR:

>>> from pathlib import Path
>>> import re
>>> for path in Path("stdlib").rglob("*.pyi"):
...     source = path.read_text()
...     new_source = re.sub(r": Literal\[([a-zA-Z0-9_\"]+)\] = \.\.\.", r": Literal[\1] = \1", source)
...     new_source = re.sub(r": None = \.\.\.", ": None = None", new_source)
...     if new_source != source:
...         path.write_text(new_source)

@JelleZijlstra we could possibly add this kind of logic to stubdefaulter, but I'm not sure if it would be a "good fit" with the rest of the code. WDYT?

[JelleZijlstra]

I'd be open to adding it.

[AlexWaygood]

I'd be open to adding it.

JelleZijlstra/stubdefaulter#54 :)