— matcher can now handle matches on edges statuses

Add method to detect if a graph is DAG (by doing a topological sort, if it fails —> cycles in the graph, otherwise, it’s a DAG)

* develop:
  Remove scripts

* 'master' of ssh://gitlab.cocophotos.eu:32423/liblinguisticgraph:

* develop:
  USE_DEPXML to avoid Xerces
  Changes for building in one time
  - lastNodeId method - know if no canonical function is expressed, don't use default final function
  Change feature method to be more extended (give any type of features now)
  - Edge properties need now a compare method to compare two edges - Graph has been updated to fit the requirement --> Now, there will be a problem, if two edges are the same in the sense of the compare method
  Changes in sugraph matching from isomorphism to monomorphism because of a bug in the VF2 algorithm implementation
  Fix bug in matcher (with edges)
  - Fix a bug on edge matching - Take lex instead of token to keep capitalization in DepXML parsing
  DepXMLParser with skip empties option New matcher for edges Regression testing in progress
  - Add checkType to check dependency edge status - New DepXMLParser (in progress)
  Small changes to find XercesC (uppercase matter !)
  - Add Xerces dependencies - Add script to list boost test - Add depxml parser - Correct minor bug in removeEdge with node's source and node's target
  Small changes in test_conll with boost::assign to stop ambiguity for G++ 4.7 on Ubuntu 13.04
  Small changes in format and graph
  Version 1.0.0 for lib LinguisticGraph
  - Better lib structure for CMake (installation is now perfect) - Add change of source or target for a node in graph.h - Add clone edge in graph.h

- know if no canonical function is expressed, don't use default final function

- Graph has been updated to fit the requirement
--> Now, there will be a problem, if two edges are the same in the sense of the compare method

Changes in sugraph matching from isomorphism to monomorphism because of a bug in the VF2 algorithm implementation

- Take lex instead of token to keep capitalization in DepXML parsing

New matcher for edges
Regression testing in progress

- New DepXMLParser (in progress)

- Add script to list boost test
- Add depxml parser
- Correct minor bug in removeEdge with node's source and node's target

- Add change of source or target for a node in graph.h
- Add clone edge in graph.h

- Lib instead of header library (because of the matcher)
- Graph, Linguistic Graph and Subgraph matching (compliant with exact parallel edges matching)