Tidy-up CUDA implementation of BwdTrans and update CUDA kernels with additional parallelism

This MR tidy-up the previous implementation of the CUDA version of the PhysDeriv operator

Operators/BwdTrans/BwdTransCUDA.cu

@@ -2,9 +2,11 @@
 
 namespace Nektar::Operators::detail
 {
+
 template <>
 std::string OperatorBwdTransImpl<double, ImplCUDA>::className =
     GetOperatorFactory<double>().RegisterCreatorFunction(
         "BwdTransCUDA", OperatorBwdTransImpl<double, ImplCUDA>::instantiate,
         "...");
+
 } // namespace Nektar::Operators::detail