Fix Qt with OpenGL on the Mac.
Add .git .bzr and .hg to the list of default CPack ignore directories.
Bump RC number
Update ChangeLog.manual

This commit cherry-picks and squashes the following commits:

  46858720 "FortranCInterface: Fix PathScale detection..." (2010-02-16)
  b39fe940 "Fix problem with ExternalProject test..." (2010-02-17)
  70290e1f "Add support for QtDeclartive module" (2010-02-17)
  282ba895 "Clarify CMAKE_MODULE_PATH documentation" (2010-02-18)
  4eba05de "Suppress GNU flag -fPIC on Windows" (2010-02-19)
  57efb4a4 "BUG: We shouldn't be setting the HideWindow..." (2010-02-19)

We store custom command rule hashes in CMakeFiles/CMakeRuleHashes.txt
persistently across CMake runs.  When the rule hash changes we delete
the custom command output file and write a new hash into the persistence
file.

This functionality was first added by the commit 'Introduce "rule
hashes" to help rebuild files when rules change.' (2008-06-02).
However, the implementation in cmGlobalGenerator::CheckRuleHashes kept
the file open for read when attempting to rewrite a new file.  On
Windows filesystems this prevented the new version of the file from
being written!  This caused the first set of rule hashes to be used
forever within a build tree, meaning that all custom commands whose
rules changed would be rebuilt every time CMake regenerated the build
tree.

In this commit we address the problem by splitting the read and write
operations into separate methods.  This ensures that the input stream is
closed before the output stream opens the file.

The --force-new-ctest-process option does not do anything in the context
where we added it, so we remove the useless change.

CTest 2.6.4 crashes if a dashboard script invokes "message()" after
"ctest_test()" or anything else that creates an inner cmCTest object.
The CMake.Install test drives installation using --build-and-test with
the outer CTest driving CMake tests.  We add --force-new-ctest-process
to avoid creation of a cmCTest object inside the outer CTest just in
case it is 2.6.4.

Commit "Add alternate per-vendor compiler id detection" on this branch
was meant for the main development line.

At least one Fortran compiler does not provide a preprocessor symbol to
identify itself.  Instead we try running unknown compilers with version
query flags known for each vendor and look for known output.  Future
commits will add vendor-specific flags/output table entries.

BUG: Fix issue #8177. Modify the system PATH for all users when requested during an NSIS based installer run. Thanks to Bart Janssens for the patch.

Make sure CTestTest scripts honor the CMAKE_TEST_GENERATOR so that if the generator to build cmake is different than the one used for tests (watcom) is used.

The commit "Fix KWSys SystemTools build on cygwin with -mwin32" tried to
restore the state of the _WIN32 definition that was broken by the commit
"Optimize KWSys SystemTools::FileExists on Windows".  It did so for the
case of building with -mwin32 on cygwin, but since including <windows.h>
defines _WIN32, it failed for the case of not using -mwin32.

This commit restores the state of _WIN32 in all cases by undefining it
after including <windows.h> if it was not defined beforehand.

Uncommented the dependency of CTestTestNoExe on CTestTestNoBuild so that it will work in parallel now.

Xcode 1.5 writes helper scripts at the projectDirPath location for
targets that do not set SYMROOT.  We now add SYMROOT to custom targets
so that all targets set it.  This prevents Xcode 1.5 from touching the
source directory now that we always set projectDirPath.

See issue #8481.

Now only the dependencies for the file where the dependencies actually may
have changed are rescanned, before that this was done for all source files
even if only one source file had changed.
This reduces e.g. on my machine the time for scanning the dependencies
of kdelibs/khtml/ when only one file (khtml_global.cpp) has changed from
around 7.5 seconds to 1.2 seconds.

The tests succeed, it does what I expected it to do on kdelibs, and Brad
also reviewed the patch, so I think it should be ok.

Alex