Distinguish on collada-export between DummyTransformationSceneNode and other nodes and export matrix for the former and position, rotation, scale for the latter. That way we can maybe get rid of the increasing the number of nodes on each import later on.

git-svn-id: svn://svn.code.sf.net/p/irrlicht/code/trunk@4274 dfc29bdd-3216-0410-991c-e03cc46cb475