Use the new large PyLong conversion functions in Py3.13.

[scoder]

Untested since it's not released yet.

See python/cpython#111140

[da-woods]

Looks fine apart from being untested.

I kind of wonder if any of this "large integer" type code is reliable covered by the test-suite at all.

[scoder]

I kind of wonder if any of this "large integer" type code is reliable covered by the test-suite at all.

This should exercise it:
https://github.com/cython/cython/blob/master/tests/run/int128.pyx

[scoder]

FAIL: signed_conversion (int128)
Doctest: int128.signed_conversion
----------------------------------------------------------------------
File "/home/runner/work/cython/cython/TEST_TMP/0/run/cpp/int128/int128.cpython-313-x86_64-linux-gnu.so", line ?, in int128.signed_conversion
Failed example:
    bigint(signed_conversion(-2**127))
Exception raised:
    Traceback (most recent call last):
      File "/opt/hostedtoolcache/Python/3.13.0-alpha.4/x64/lib/python3.13/doctest.py", line 1378, in __run
        exec(compile(example.source, filename, "single",
        ~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                     compileflags, True), test.globs)
                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
      File "<doctest int128.signed_conversion[19]>", line 1, in <module>
        bigint(signed_conversion(-2**127))
               ~~~~~~~~~~~~~~~~~^^^^^^^^^
      File "tests/run/int128.pyx", line 139, in int128.signed_conversion (int128.cpp:2833)
        cdef int128_t n = x
    OverflowError: value too large to convert to __int128_t


======================================================================
FAIL: unsigned_conversion (int128)
Doctest: int128.unsigned_conversion
----------------------------------------------------------------------
File "/home/runner/work/cython/cython/TEST_TMP/0/run/cpp/int128/int128.cpython-313-x86_64-linux-gnu.so", line ?, in int128.unsigned_conversion
Failed example:
    bigint(unsigned_conversion(2**128-1))
Exception raised:
    Traceback (most recent call last):
      File "/opt/hostedtoolcache/Python/3.13.0-alpha.4/x64/lib/python3.13/doctest.py", line 1378, in __run
        exec(compile(example.source, filename, "single",
        ~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                     compileflags, True), test.globs)
                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
      File "<doctest int128.unsigned_conversion[19]>", line 1, in <module>
        bigint(unsigned_conversion(2**128-1))
               ~~~~~~~~~~~~~~~~~~~^^^^^^^^^^
      File "tests/run/int128.pyx", line 75, in int128.unsigned_conversion (int128.cpp:2670)
        cdef uint128_t n = x
    OverflowError: value too large to convert to __uint128_t

[scoder]

I found several ref-leak issues in the shift+mask fallback conversion implementation while working on this that should also be fixed in 3.0.x. It's probably easier to backport the PR than to try to extract them separately.