New Features:

* Graphical simulation result is displayed in a table view to show
  run-time and overlap information if the corresponding processors
  are loaded. The Simulation result dialog can be deactivated in
  the preferences.
* PedestrianOverlapProcessor returns two values "distance", "overlaps"
  for each overlap detected. If no overlap occurs the output is empty.
  "distance": The distance between the center of the two pedestrians
  "overlaps": The amount the two pedestrian overlap

Changes:

* New version for scenario.json (0.5):
  - remove scenario/attributesSimulation/needsBoundary as it is not
  used anymore.
  - remove attributes from processor PedestrianOverlapProcessor because
  the radius of a pedestrian can be retrieved from the topography at
  run-time. This would allow to detect overlaps of different sized
  pedestrian in the future.
* Helper interfaces for testing:
  - The TestResourceHandler interface groups all methods
  needed to retrieve backup and reset test resources between single test
  runs. All methods, except getTestDir() have default implementation
  which rely rely on relative paths on getTestDir().
  - The JsonNodeTest interface groups all methods needed to travers or
  search the Jackson JsonNode data structures. In contains methos for
  path finding and will fail the test if a path does not exist.
* Added Test for PedestrianOverlapProcessor to check if a overlap
  of (2*r - 0.001) will always be treated as an overlap. Additional
  test are  created to ensure (2*r + 0.001) with r:= pedestrian radius,
  will never create an overlap. Only between
  [2*r - 0.001 , 2*r + 0.001] different placements of pedestrian can
  produce overlaps. Example: --,  /(45 degree) , |  represent the
  distance (2*r)

  	Case A:		Case B: 	Case: C
	   O--O		  O 		  O
	   			  |			   \
	   			  O             O

  In the Case A and B no overlap is detected. In Case C an overlap
  is detected, because of rounding errors when calculating the
  distance \.
  This does not occure if the overlap is bigger than 0.001. Then in all
  three case a overlap is detected. If the distance between two
  pedestrian is 0.001 bigger than 2*r in none of the above cases an
  overlap is detected.

# Conflicts:
#	VadereModelTests/TestOSM/scenarios/corner_waiting_time_processor_test.scenario
#	VadereModelTests/TestOSM/scenarios/narrow_passage_discrete_CA_fail.scenario

# Conflicts:
#	VadereModelTests/TestOSM/scenarios/corner_waiting_time_processor_test.scenario
#	VadereModelTests/TestOSM/scenarios/narrow_passage_discrete_CA_fail.scenario

Renamed four TestSFM RIMEA scenarios which includes too many persons to run in a reasonable time. Files are renamed to *.backup instead of *.scenario.

update PSO such that it uses more iteration to improve accuracy. Update OSM-ModelTests: add TestProcessors, update attributes of the tests. add additional scenarios.

- PedestrianOverlapDistProcessor
  Saves for each pedestrian in each timestep the size of the overlap. That is the difference between the radii of the pedestrians and the distance between their centers. The processors works only if all pedestrians in the simulation have the same radius!
- MaxOverlapProcessor
  Saves the maximum overlap for one simulation. As explained above, the overlap is 2*pedRadius - norm(ped_pos_i - ped_pos_j). Again, this processor only works if all pedestrians in the simulation have the same radius.

Both processors are especially useful for analyzing the number and sizes of overlaps occuring in the simulation. Both processors were tested with a Matlab script but there are no JUnit tests yet.

implementation of some TestProcessors, update of OSMTests e.g. add the TestProcessors to the scenarios.

Tests SFM: Activated useFreeSpaceOnly for spawning the pedestrian in the sources. Added necessary output processor for rimea_06_corner_sfm1.scenario (due to dependency pedestrianStartTimeProcessor was needed). Reduced the number of agents in basic_1_chicken_sfm*.scenario from 200 to 50 to reduce the computation time.

increase size of spwan grid in SourceController to ensure peds don't overlap