Re-enable windows backends in autotool

Closes #542

See merge request cairo/cairo!291

AC_LANG_SOURCE only takes one argument, ignored the others.

Closes #542

build: Remove unused DirectFB bits on Windows

See merge request cairo/cairo!288

pdf: ensure urls and filename strings are correctly encoded

Closes #526

See merge request !266

The DirectFB surface removal commit did not remove all the build bits.

quartz: Fix spelling in comment

See merge request cairo/cairo!286

Remove old backends

See merge request cairo/cairo!285

quartz: Improve use of images for better performance.

See merge request cairo/cairo!282

The CI runners for Windows are currently timing out on the fd.o
infrastructure. The issue is being worked on, but we should avoid
blocking our entire CI pipeline until it's fixed, as it sounds it is
going to take a while.

OpenVG is a failed experiment from 10 years ago, and nobody has been
using the cairo-vg surface in pretty much the same amount of time.

Nobody cares about DirectFB any more, not even the people that registered
and maintained the DirectFB domain.

Cogl's upstream is unmaintained, and has been for the past 6 years.

The DRM backend has always been a science experiment, but now it hasn't
been built in more than 10 years, and it's completely broken.

OS/2 support was last built in Cairo 1.12, which was released 10 years
ago.

Additionally, OS/2 is not supported by Meson.

The Skia surface is not there any more.

Nobody cares about BeOS in 2022, and it cannot be built with Meson anyway.

Removing the BeOS backend also allows us to drop the dependency on a C++
compiler and simplify the Autotools build considerably by removing all
the extra CXX targets.

It's disabled by default, and unsupported as it depends on Qt4, which
has been EOL since 2015.

CGContextDrawTiledImage turned up as a significant time-user while
profiling a benchmark created to evaluate
https://gitlab.gnome.org/GNOME/gtk/-/issues/3714. Without this commit
the benchmark is able to perform a mean frame rate of 2.19 frames per
second with a standard deviation of 0.09; with the commit the mean frame
rate is 2.37 fps, s.d. 0.30, both over 15 10-second samples. Student's
t-test reports a 9.8% likelyhood that the two represent the same
distribution.

To simplify _cairo_quartz_setup_state.

Caching doesn't really do anything and removing it provides a 50%
speedup and gets pdf-operators-text to pass on argb32.

Motivation: Avoid need to recreate CGImages for unchanged surfaces,
an expensive operation, while ensuring that the CGImages are properly
freed and new ones created when the surface does change.

Thanks to Uli Schlacter for suggestion and coding guidance.

Fix integer overflows in cairo-mempool

Closes #510

See merge request cairo/cairo!253

test/hairline.c: Use Unix line endings & add copyright header

See merge request cairo/cairo!234

quartz: Fix spelling in comment

See merge request cairo/cairo!284

miscellaneous math fixes

See merge request cairo/cairo!280

In a couple of instances, ints are being casted to int before use.

In a couple of instances, code is present where two numbers are being
multiplied in a type like unsigned int, but immediately being casted
to a wider type like size_t.

This means, although the result can be any size_t value, the
multiplication can potentially overflow before it's used because
unsigned int has a smaller range of values.

In another more niche case, I also cast to size_t before multiplying
a signed integer, since the result is immediately used as an argument
to memcpy, which would give memory corruption if the value was negative
anyway.

added missing documentation to cairo_font_options_status

See merge request cairo/cairo!281

The documentation for cairo_font_options_status says it can return
either CAIRO_STATUS_SUCCESS or CAIRO_STATUS_NO_MEMORY, but the
implementation can also return CAIRO_STATUS_NULL_POINTER, which wasn't
mentioned.

In some places, there were int variables being compared to unsigned
ints when they would never take a negative value, exposing some edge
cases that didn't need to be there.

meson: Add xml option and disable by default

See merge request cairo/cairo!275

Just like autotools does.

[user-font] Implement subpixel-positioning

See merge request cairo/cairo!274

See:

https://github.com/harfbuzz/harfbuzz/pull/3411#issuecomment-1033118545
https://github.com/harfbuzz/harfbuzz/pull/3411#issuecomment-1033125311
https://github.com/harfbuzz/harfbuzz/pull/3411#issuecomment-1037765840

Basically the code in cairo-image-compositor.c:composite_glyphs()
is flooring glyph positions and putting some phase in the glyph-id.
This was being lost in the user-font backend. Fetch that and put it
in the recording-surface's device-transform translation, to be applied.

Fix mixed color clusters

See merge request cairo/cairo!273