Export the cholmod_factor object

See merge request core/dune-istl!455

This is needed for some of the more advanced features of CHOLMOD.

refactor the way python is used in dune

See merge request core/dune-istl!432

comment out CMP0087 NEW cmake policy because it fails on some CI runners

correctly set CMP0087 cmake policy to NEW for cmake >= 3.14

fix ci

pep 42: use implicit namespace packages which make it possible to combine subpackages using editable (source) and non editable (pypi)

only use a cmake install a python module in the SKBUILD setting - for an install from build dir pip will take take of this since it is part of the setup.py file

add a stopgap solution unti ci repo up and running again

Update requirements.

add dune packages to requires in setup.py.in

[testing]

only add dune modules that have been found as requirements in setup.py.in

removed setup.py.in since the default is provided in dune-common

adapt ci yml file to python bindings on by default

disable python bindings for clang 5 (ubuntu 18) since no numpy matching numpy version can be installed from the gitlab registry

[bug fix python bindings]
fix an issue with 'entry' being used in implicit build mode but ready==build

[minres] Compute preconditioned residual also for initial residual

See merge request core/dune-istl!454

Bump version to 2.9-git

See merge request core/dune-istl!451

Make sure that the relaxation factor for smoothers never is complex.

See merge request core/dune-istl!447

Remove signed-unsigned comparison on repartition.hh

See merge request core/dune-istl!448

to get rid of a compiler warning.

The note

/dune-istl/dune/istl/paamg/smoother.hh:130:26: note: the ABI of
passing structure with complex float member has changed in GCC 4.4

for DefaultConstructionArgs::getArgs got me thinking. I just could not
imagine why there should be comlex value involved. Turned out we
somehow assumed that for complex the relaxation factore would be
complex, too. IMHO a relaxation factor should scale and hence be real.
This MR make ssure that the relaxation factor is never complex.

Unforntunately, then a note for std::make_pair with a
complex in repartition.hh appeared:
/usr/include/c++/8/bits/stl_pair.h:524:5: note: the ABI of passing
structure with complex float member has changed in GCC 4.4
     make_pair(_T1&& __x, _T2&& __y)

and I just resorted to constructing Pair directly. Not sure whether
this is really needed but it should not hurt and ensures more silence.

Get rid of shadowed/unused variable/typedef gcc-8 warnings

See merge request core/dune-istl!445

Get rid of unused/shadowed variable compiler warnings.

See merge request core/dune-istl!442

[gitignore] Ignore macOS caching files

See merge request core/dune-istl!439

[MultiTypeBlockMatrix] Export the constructor of std::tuple

See merge request core/dune-istl!395

[changelog] Mention SuiteSparse potentiel slowdown with openBlas

Closes #91

See merge request core/dune-istl!437

Fixes #91

Update list of authors

See merge request core/dune-istl!438

Cholmod: Add support for almost all matrix types

See merge request core/dune-istl!424

This is achieved by using the flat matrix and vector forEach implementations.
The unit test is extended by a multi type example.

The routines traverse the container recursively and call a functor
at each scalar entry passing also the flat index.

Remove some warnings about unused typedefs and suggested braces

See merge request core/dune-istl!435

IO matrixmarket for vectors of simd type

See merge request core/dune-istl!368

[operators][cleanup] Use override keyword

See merge request core/dune-istl!433

use the dune docker registry for the python and 'with-checking' ci test

See merge request core/dune-istl!431

fix some headercheck issues

See merge request core/dune-istl!430