bpo-32492: Tweak _collections._tuplegetter.

[serhiy-storchaka]

• Remove the docstrings cache.
• Improve tests.
• Unify names of tp_descr_get and tp_descr_set functions.

https://bugs.python.org/issue32492

[rhettinger]

Consider adding another test for inspect.isdatadescriptor(Point.x)

[rhettinger]

If caching is taken out, then docstring interning should be added back.

Python 3.8.0a0 (heads/pr_11367:906c21fb8b, Dec 30 2018, 14:49:20)
[Clang 10.0.0 (clang-1000.11.45.5)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> from collections import namedtuple
>>> P = namedtuple('P', ['m', 'n'])
>>> Q = namedtuple('Q', ['o', 'p'])
>>> P.m.__doc__ is Q.o.__doc__
False
>>> P.m.__doc__ == Q.o.__doc__
True

[rhettinger]

Also, the test for the exact help() string seems fragile (it depends on the exact spacing and formatting of the help() output). Instead, just check for the presence of the docstring or check that it is data descriptor and leave it at that.