CUTLASS example added, license headers added, fixes

- Add license header to each example file.
- Fixed broken runs caused by type declarations.
- Fixed hang in throughput.py when --run-once by doing a
  manual warm-up step, like in auto_throughput.py

python/examples/requirements.txt

@@ -0,0 +1,7 @@
+numpy
+numba
+cupy
+nvidia-cutlass
+cuda-cccl
+cuda-core
+cuda-bindings