Optimizations: global body intersections and device transfers

examples/modular/sd_1.cpp

examples/modular/sd_2.cpp

examples/modular/sd_3_CPU.cpp

examples/modular/sd_3_GPU.cu

examples/modular/sd_4.cpp

runtime/pairs.cpp

src/pairs/code_gen/cgen.py

src/pairs/code_gen/interface.py

src/pairs/ir/kernel.py

src/pairs/ir/loops.py

src/pairs/mapping/funcs.py

src/pairs/sim/global_interaction.py

src/pairs/sim/interaction.py

src/pairs/transformations/devices.py