GitLab

ParallelTTracker. Particle-Matter interaction doesn't work either at the moment.

I hope that I didn't create too much of a mess when merging!

remove particle with ID==0 form the H5 file in case of opal-cycl and from the statistics calculation

moving Classic into src directory and pack all objects into libOpal since Classic also depends on libOpal

modified:   classic/src/Algorithms/PartBunch.h

Code cleaning:
	The following in OPAL unused functions and variables of PartBunch are removed:

	void stash()
	void pop()
	Vector_t getStashIniP() const
	UnitState_t getUnitState() const
	void makHistograms()
	double getRho(NDIndex<3> e)
	bool itIsMyTurn(int *n)
	bool hasZeroNLP()
	void set_nBinsLineDensity(int n)
	void setGridIsFixed()
	double getRebinEnergy()
	void weHaveNOBins()
	double calcTimeDelay(const double &jifactor)
	void moveBunchToCathode(double &t)
	void printBinHist()
	void updatePartInBin(size_t countLost[])
	void updateBinStructure()
	void reportParticleLoss()
	double getMaxdEBins()
	virtual void setKR(Vector_t value, int i)
	virtual void setKT(Vector_t value, int i)
	virtual void BetOut(FILE *dat, FILE *sli)
	virtual void plotR()
	void beamEllipsoid(FVector<double, 6>   &centroid, FMatrix<double, 6, 6> &moment)
	void calcWeightedAverages(Vector_t &CentroidPosition, Vector_t &CentroidMomentum) const
	bool isZPeriodic() const
	double getGaBeLa() const
	double getZPos()
	void getXBounds(Vector_t &xMin, Vector_t &xMax)
	double   get_phase() const
	double* get_energy()
	void set_hr(Vector_t h)
	double get_meshEnlargement() const
	size_t getMinLocalNum()
	void calcBeamParametersLight()
	inline bool hasLowParticleCount()
	size_t getNumPartInBin(int BinID) const
	bool resetPartBinID()
	double getTBin()
	bool isDcBeam()
	Field_t tmpFieldZ_m
	Vector_t csBeta_m
	Vector_t csAlpha_m
	Vector_t stash_iniP_m

Added P3M solver for periodic domain. Introduce disorder induced heating test in a constant focussing channel. This test is exectued when selecting P3M as solver.

Workaround to run the degrader and not break the regressiontests. This is TEMPORARLY until we have understood the problem and At THE SAME TIME satisfy our users

- Expressions: adding error function
- LossDataSink: we are using git
- OpalBeamline: let Autophase tracker smoothly (i.e. don't throw a segmentation fault) even if there are no cavities
- ParallelTTracker: don't save phase space after last step if saved at end of last step

  data array
- PartBunch: slightly enlarge mesh to prevent that round-off errors
  result in access to memory that is not allocated

- new laser profile class with automatic detection of center of mass of profile and possiblity to flip and rotate

 - reused code
 - properly freeing memory of gsl_rng structure
 - initializing members