refactor: root kernel includes at csrc/kernels

- add the kernels directory to CMake target include paths and drop all ../-relative includes in kernel sources
- reference shared primitives as common/*.cuh and the fp8 type header as fp8/common.h
- update standalone test nvcc commands in file headers and cuda_kernels.md to -I csrc/kernels

csrc/CMakeLists.txt

@@ -90,6 +90,7 @@ foreach(i RANGE ${_kernel_last})
     target_compile_definitions(${name} PRIVATE TORCH_EXTENSION_NAME=${name})
 
     target_include_directories(${name} PRIVATE
+        "${CMAKE_CURRENT_SOURCE_DIR}/kernels"
         "${TORCH_HOME}/include"
         "${TORCH_HOME}/include/torch/csrc/api/include"
         "${PYTHON_INCLUDE_DIR}")