* Updated CMake scripts to work with new CMake 3 and TBB 4.3

* Introducing visual regression testing framework
* Added PipelineResultImageTest case to run pipelines and write their rendering results to PNGs.
* Updated DevilImageWriter, added property to make writing depth images optional.

1. edited test-campvis::main to generate xml regardless of environment var. 2. check is the result image null. 3. open and edit the gtest output xml with results from python test

Registered testing thread and OpenGL context with OpenGLJobProcessor.

Cleaned up DevilImageReader to be working with all kinds of input images and being completely OpenGL free.

* Updated DevilImageWriter, added property to make writing depth images optional.

Fixed GenericImageReader referencing DevilImageReader even if the corresponding module was not present.

* AbstractImageReader is now again an AbstractProcessor, so that we can get rid of the dynamic_casts in GenericImageReader
* DevilImageReader is no longer a VisualizationProcessor but gets the OpenGL context through OpenGLJobProcessor::ScopedSynchronousGlJobExecution
* Fixed alglib's CMakeLists.txt

3d mi segment flat rebased

See merge request !89

campvis-test does not use OpenGLJobProcessor and only initializes it halfway. However, some CAMPVis functionality relies on that. Therefore, I introduced a hack/workaround to set the OpenGL thread manually if you need to.

Furthermore, added exit(EXIT_FAILURE) to recursive assertions to avoid possible endless assertion loops in campvis-test (and thus amok running jobs in Jenkins).

* Fixed possible endless loop in event evaluation of AbstractPipeline.
* Fixed 4-channel image rendering in datacontainerinspector.frag.

* Added OpenGL thread guard to image conversions to ImageRepresentationGL.

[merged commit]clean ups, fixes for g++ errors and warnings, and edits for cppcheck errors (except for the 480+ in the external library AlgLib)

Various fixes

See merge request !91

The VolumeExplorer has now an additional property defining which view is shown in the big section (and which in the three small sections). Hence, now it is also possible to show one of the slice views in large and have the volume rendering in small.

To better support sharing the functionality of slice rendering, the SliceExtractor processor was refactored: Similar to the RaycastingProcessor, the main functionality was moved to an abstract base class, the SliceRenderProcessor. This takes care of computing all the necessary transformation matrices, optionally rendering crosshair and integrating geometry, as well as handling user input to support scribbling. The actual SliceExtractor processor now only implements the rendering of the slice itself, i.e. applying the transfer function.

In this regard, VolumeExplorer was updated to use the generic SliceRenderProcessor, and TransFuncWindowingEventListener supports changing of its assigned property.

Replaced DataContainer::s_dataAdded<const std::string&, DataHandle> with DataContainer::s_dataAdded<std::string, DataHandle>.

initial py script. given 3 directory compares reference and testruns to store results

According to Visual Leak Detector, CAMPVis is now memory leak free. :)

Memory pools

See merge request !90

* Fixed includes in sigslot.cpp
* Implemented memory pools also for AbstractJob using the pool of OpenGLJobProcessor