Problem 691

As report#690, Hadron Physics of TestEM7 has a problem.
This problem is produced by the following way.

Some Particles created in Binary Cascade ( and also Binary Cascade Light Ion)
do not have ProcessManger.
So that at the applying DecayProcess for all applicable particles in
PhysicsList::ConstructProcess,
pmanager ->AddProcess(fDecayProcess); line makes core dump,
because pmanger is NULL.

The simplest way to avoid this trouble is to change the order of preocess
creations. To apply DecayProcesses ahead to the creation of Hadronic Processes.

Change PhyisicsList.cc ConstructProcess() like this,
*******************************************************************************
  // decay process
  //
  G4Decay* fDecayProcess = new G4Decay();

  theParticleIterator->reset();
  while( (*theParticleIterator)() ){
    G4ParticleDefinition* particle = theParticleIterator->value();
    G4ProcessManager* pmanager = particle->GetProcessManager();

    if (fDecayProcess->IsApplicable(*particle)) {
      pmanager ->AddProcess(fDecayProcess);

      // set ordering for PostStepDoIt and AtRestDoIt
      pmanager ->SetProcessOrdering(fDecayProcess, idxPostStep);
      pmanager ->SetProcessOrdering(fDecayProcess, idxAtRest);
  }

  // hadronic physics lists
  for(size_t i=0; i<hadronPhys.size(); i++) hadronPhys[i]->ConstructProcess();


*******************************************************************************

The other way is to check a particle has a processManager or not in applying
the Decay Processe.

The study for more essential solution may be necessity.

Tatsumi