Drop dependency on decorator module

See merge request inducer/pycuda!53

Update mempool to version from PyOpenCL

See merge request !52

Add links to related software

See merge request !50

Bump size parameters to size_t for a few memcpys

See merge request !49

Remove py-version-specific Trove classifiers from setup.py

See merge request inducer/pycuda!48

Cuda array interface

See merge request !44

CI/README: master->main

See merge request !46

Add a pyproject.toml to control build-time numpy version

See merge request !45

Build against the oldest C_API_VERSION available for the current Python
version, to maximise compatibility of the resulting installation.

Closes #271.

This commit adds support for version 3 of the CUDA Array Interface
(https://numba.readthedocs.io/en/latest/cuda/cuda_array_interface.html):

- GPU Arrays export the interface so that they can be passed to other
  supporting libraries.
- Function calls check for the presence of the interface and use the
  data pointer from it if found, so other libraries' arrays can be
  passed to PyCUDA-compiled kernels.

Documentation, tests, and examples for the interface going in both
directions are added.

Upgrade compyte for sized bool

See merge request !43

Add documentation memset_*_async.

Closes #266.

add "out" parameter to GPUArray.conj(); add equivalent GPUArray.conjugate() method

See merge request !42

add equivalent GPUArray.conjugate() method

Check contiguity of arrays in GPUArray.fill (closes gh-265)

See merge request !41

Shorten/standardize doc/conf

See merge request !40

Drop usage of six

See merge request !39