Add experimental half precison support

With this MR experimental half-precision support is added

pystencils/gpu/__init__.py

 from pystencils.gpu.gpu_array_handler import GPUArrayHandler, GPUNotAvailableHandler
-from pystencils.gpu.cudajit import make_python_function
+from pystencils.gpu.gpujit import make_python_function
 from pystencils.gpu.kernelcreation import create_cuda_kernel, created_indexed_cuda_kernel
 
 from .indexing import AbstractIndexing, BlockIndexing, LineIndexing