a correct instalation of Python in OSX

update packaging script to include Python in deployed DMG

In case of conflict in i18n, keep the version of the branch "ours"
To resolve a particular conflict, "git checkout --ours path/to/file.desktop"

The patch fixes two problems:

1) DlgAnimationRenderer didn't initialize frames' mimetype correctly,
   so exporting always happened into PNG format.

2) DlgAnimationRenderer should correctly initialize saving format
   settings from the default cfg.exportConfiguration(). Using the
   same settings for both, animation rendering and normal image save
   is a disputable approach, but that  is how we do it for ages.

3) EXRConverter shouldn't call any blocking functions on the image
   (and shouldn't change the state of the image as well). So it should
   convert non-16f/32f layers on the fly.

BUG:406830

It seem to be a general convention in other applications, like
Firefox and QtSVG. The main purpose of such estimation is to
read correct aspect ratio.

CCBUG:406124

It's optional for gmic and gmic-qt, but CImg.h demands fftw_thread_init
unconditionally.

This built for me, on my test vm...

Maybe this'll inspire people to check Jenkins more often...

This adds a series of patches related to QTBUG-53022 [1] for Windows to
allow enabling support for fractional DPI scaling.

These series of patches adds the option
`QGuiApplication::setHighDpiScaleFactorRoundingPolicy` to control how
the DPI scale factor is rounded (or not), which can also be overridden
by the environment variable `QT_SCALE_FACTOR_ROUNDING_POLICY`. The
options include:

- Round: Round up for .5 and above.
- Ceil: Always round up.
- Floor: Always round down.
- RoundPreferFloor: Round up for .75 and above.
- PassThrough: Don't round.

Currently the default option is explicitly set to `Round` to obtain the
old behaviour (where 1.5x is rounded to 2x).

Ref:
- https://bugreports.qt.io/browse/QTBUG-53022
- https://codereview.qt-project.org/157173
- https://codereview.qt-project.org/157174
- https://codereview.qt-project.org/161334
- https://codereview.qt-project.org/163219
- https://codereview.qt-project.org/176381
- https://codereview.qt-project.org/209886

This adds a patch for Windows to fix moving windows between screens with
different DPI scaling.

Ref: https://codereview.qt-project.org/248822

This avoids drawing the path twice on load, update is made only from setFilter
which is also called on document load. If done from the creator it draws
the default pattern first and then overlays the saved pattern.

BUG:408169

Every KoShapeController used to reset the global "shape document" in its
KoDocumentResourceManager. The problem is that in Krita we have multiple
KoShapeController-s, but the resource manager is the only one. It means
that every shape selection or shape layer used to reset the pointer to
the global "shape document" to its own one. The only reason why we didn't
have too many crashes is that almost noone uses this "shape document".

Ideally we should:
1) Remove KoDocumentResourceManager::GlobalShapeController resource. This
   resource is not global to the document, but is created per-layer. The only
   user of it atm is the legacy text tool, which should be deprecated.

2) Remove KoDocumentResourceManager at all. There is no much use of it
   anymore. Its main user is legacy text tool.

BUG:407554

Before this commit, some percentages of scale would result
in the content disappearing after using Transform Tool.
Adding more support to the filter strategy in this commit
ensures that the correct pixel is added to the pool of possible
pixels for NN algorithm.

BUG:408182

Before this commit, there was only a few tests for NN algorithm,
this commit adds more of them, including a series of tests for
all % of scale from 0 to 100.

This makes it a tiny bit easier to see which selection option
is selected.

BUG:408343

[171/4490] Building CXX object libs/odf/tests/CMakeFiles/TestStorage.dir/TestStorage.cpp.o
/home/boud/dev/krita/libs/odf/tests/TestStorage.cpp: In member function ‘void TestStorage::storage()’:
/home/boud/dev/krita/libs/odf/tests/TestStorage.cpp:85:15: warning: ignoring return value of ‘int system(const char*)’, declared with attribute warn_unused_result [-Wunused-result]
         system(ba.constData());       // QDir::rmdir isn't recursive!
         ~~~~~~^~~~~~~~~~~~~~~~
/home/boud/dev/krita/libs/odf/tests/TestStorage.cpp: In member function ‘void TestStorage::storage2()’:
/home/boud/dev/krita/libs/odf/tests/TestStorage.cpp:218:15: warning: ignoring return value of ‘int system(const char*)’, declared with attribute warn_unused_result [-Wunused-result]
         system(ba.constData());       // QDir::rmdir isn't recursive!
         ~~~~~~^~~~~~~~~~~~~~~~

This apparently fixes an issue with duplicate targets that Ninja
has decided to error out on.

BUG:407347

BUG:408344

This reverts commit 75df7ccf.

This doesn't apply at the moment and needs looking into again.