One of the latest patches prevent the macros of the current module
form being executed. This patch fixes this by including the module
in the dependency tree.

When making the entries of ALL_DEPENDENCIES unique, we skipped modules that
did not ship a Dune<ModuleName>Macros.cmake macro. This caused their libraries
to be missing in DUNE_LIBS. This patch fixes this situation.

Previously we already processed dune.module but in a much later stage during
configure. Dependencies are contained in dune.module. Therefore there is no
need to replicate them in the CMake package configuration files. With this
patch we now always process dune.module directly after finding the DUNE module.
In addition we set ${DUNE_MOD_NAME}_(DEPENDS|SUGGESTS} in the CMake environment.
As with this change DUNE_MOD_NAME will usually be set to to dune-common, we
also changed to using ProjectName and ProjectVersion (the canonical CMake names)
to refer to the current module and version.

In the CMake package configuration files generated with autotools
the CMake name of the library target contained a bogus lib in front
of it. This packages fixes this. Now the library can be referenced
in CMake as dunemodule (without dashes).

Before this patch we tried setting am_dir for directory with a
DUNE module. If one is working with local modules
(e.g. dune-common and dune-geometry), that are a subset of installed
modules (e.g. dune-common, dune-geometry, and dune-istl), that are all
installed under the same prefix, then modules that are part of the difference
of the set of installed and the set of the local modules caused am_dir to
point to the am directory of the installed dune-common directory.
After this patch  am_dir is only set once, which fixes this behaviour and
flyspray issue #1420

Fixes flyspray issue 1418, see https://dune-project.org/flyspray/index.php?do=details&task_id=1418

when we try to send data, it might happen, that for some reason we don't want to any data.
This happens, when the dynamic size of our CommPolicy is 0. In that case we must not add a
MessageInformation, otherwise we get an MPI error.

Somewhere along the many changes we stopped forcing usage of pkg-config when
testing for installed versions of modules during configure. Without using
pkg-config there is no way of detecting custom installation parts used by
the user. In my case, this broke finding modules installed via cmake from
modules that are to built with autotools.

This patch forces DUNE_CHECK_MODULE to use pkg-config to setup the paths for
installed modules by dunecontrol not provinding --with-module=<path> to configure
for installed modules.

Improve comments.

Previously we checked whether the extracted Version evaluated to
True. Unfortunately, there lazy people like me that like use a version
of 0 for testing, which evaluates to False and breaks configure.
With this patch we allow 0 to and just test wether the version consists
of characters.

A variable was not correctly escaped during creation of the file
cmake/modules/Makefile.am causing the error message:
"duneproject: line 914: datadir: command not found."

This patch now correctly escapes it.

duneproject failed miserably when run with only installed modules. With
this patch we correctly setup the pkg-config and dunecontrol path to
handle this situation. In addition we fixed various issues with the newly
created modules and they now build as exspected.

With new modules there were cases where cmake/modules was not in the
modules path which made configuration fail. This patch fixes this case.

Due to testing the wrong directories some variables where missing in the
package configuration files. This patch fixes this.

The clutter the output.  I can't believe anyone really wants to see
the friend relationships in doxygen.

Somehow, the vector version of CollectiveCommunication< MPI_Comm >::prod
passed std::plus to allreduce (instead of std::multiplies). This patch
corrects this flaw.

Previously some of the pathes (e.g the multiarch ones) took precedence
over the others. But as with same some of the pkgconfig files are installed
below lib/<multiarch> and others (if there is not object library) are not.
With this patch all possible install locations of pkgconfig files are added
to the path if they exist.

With openSuse 12.3 (Doxygen 1.8.5) the generated documentation contains
lots of huge graphs which show the include dependency graph of a file.
This lead to large tarballs of 155 MiB compared to 50 MiB with this
patch.
With Debian Wheezy (Doxygen 1.8.1.2) the tarballs did not blow up in
file size. Scanning the Doxygen change log did not enlighten me and
I remain puzzled why the graph were never built.

INCLUDED_BY_GRAPH is changed to prevent the same behvior for future
releases. We currently do not include these graphs and this change
makes the configureation consistent with the result.

The compiler might exit with nonzero exit code if given non-existent
options. Therefore we need to use set +e during this to make dunecontrol
to continue running even the options are not supported.

Some ^[$BLANK] or [^[$BLANK]^'], where $BLANK is space followed by tab
did not work as a capturing statement in some cases (e.g. opts files
with CONFIGURE_FLAGS="CXX='bla'"). After this patch we use [$NOBLANK]
with NOBLANK=^$space^$tab which works. In addition the other patches
form feature/posix-sed-dunecontrol have been backported to close flyspray
1401.