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/kernels/attention/decode_split_kv.cuh

@@ -2,8 +2,8 @@
 #include <cuda_bf16.h>
 #include <float.h>
 #include "common.h"
+#include "common/reduce.cuh"
 #include "layout_policies.cuh"
-#include "../common/reduce.cuh"
 
 namespace astrai {
 namespace attention {