Property values can be legitimately missing when the values have never
been changed from the default ones.

Change-Id: I3cfb6fdb41ce0012ee8af61e0566d2f8b5df78e3

Incidentally, this fixes the dbgutil error problem.

The old code was passing a copy of the ScHeaderFooterContentObj object
to the ScHeaderFooterTextData object, which was why the counter was
incremented on copy (though they should still have used a wrapper instead
of manual reference counting).  My new code passes a reference, which
removes the copying, but those pesky acquire() and release() calls
were still there.  This caused the ScHeaderFooterContentObj object to
never get deleted, which in turn caused the dbgutil refcounting check
of SvxUnoTextRangeBase object to fail.

Change-Id: I4e1780b32b21098a121ec7a43bb3eb842a419584

Change-Id: I30c370f0479743a831be29a0fa7cb447c7119ee4

LibreOffice has now been packaged in pkgsrc-wip, a DragonFly-specific
configuration file doesn't make sense.

 # Changes to be committed:

...that are no longer bundled extensions.  Otherwise, old user data where these
were still recorded as bundled extensions could cause execeptions at start up
about duplicate implementations.

...detected now that the default service manager XContentEnumerationAccess
no creates its own XServiceInfo instances.

...otherwise, if they are bundled, their per-user data is not regenerated, leading
to inconsitencies.

The tests themselves do pass, however.  The crash is caused by some
sort of cleanup, due to someone calling a pure virtual function or
some such.

I'll work on converting these tests into C++.

Thereby laying down the foundation for arbitrary field support in cells.

This cleans up the code quite a bit.