2. particles are allocated on GPU
3. GetI()/SetI() are adapted for GPU
4. added GetAllPartNumGPU(), GetNewParticleGPU(),
DeleteParticle_withoutTrajectoryTerminationGPU()r
5. modified: LapentaMultiThreadedMoverGPU(),
ExchangeParticleData(),PIC::ParticleBuffer::Init()
6. PIC::ParticleBuffer::Thread is declated _TARGET_DEVICE_
_CUDA_MANAGED_ now
7. test/srcFastWave/main.cpp: particles are initialized on GPU

test: test_fast-wave passes