Bring some optimisations for single GPU

.gitignore

HySoP/hysop/.pyflymakercc

HySoP/hysop/gpu/cl_src/advection/basic_rk2.cl

HySoP/hysop/gpu/cl_src/advection/basic_rk2_noVec.cl

HySoP/hysop/gpu/cl_src/advection/basic_rk4.cl

HySoP/hysop/gpu/cl_src/advection/basic_rk4_noVec.cl

HySoP/hysop/gpu/cl_src/advection/builtin_rk2.cl

HySoP/hysop/gpu/cl_src/advection/builtin_rk2_noVec.cl

HySoP/hysop/gpu/cl_src/advection/builtin_rk4.cl

HySoP/hysop/gpu/cl_src/advection/builtin_rk4_noVec.cl

HySoP/hysop/gpu/cl_src/advection/comm_basic_rk2_noVec.cl

HySoP/hysop/gpu/cl_src/advection/comm_builtin_rk2_noVec.cl

HySoP/hysop/gpu/cl_src/advection/velocity_cache.cl

HySoP/hysop/gpu/cl_src/advection/velocity_cache_noVec.cl

HySoP/hysop/gpu/cl_src/common.cl

HySoP/hysop/gpu/cl_src/kernels/advection.cl

HySoP/hysop/gpu/cl_src/kernels/advection_and_remeshing.cl

...hysop/gpu/cl_src/kernels/advection_and_remeshing_noVec.cl

...p/gpu/cl_src/kernels/advection_and_remeshing_vector_2d.cl

...p/gpu/cl_src/kernels/advection_and_remeshing_vector_3d.cl