Update CUDA to SYCL converter naming (codeplaysoftware#138)

The folder names have been updated, and minor corrections made to the text of the README.

Author: Graham-Codeplay

utils/cuda_to_sycl/convertor/README.md renamed to utils/cuda_to_sycl/converter/README.md

@@ -1,14 +1,17 @@
-# The SYCL Convertor
+# The SYCL Converter
 
-A `*.cu` file is a combination of Host CUDA API and device CUDA kernel.
+This utility enables developers with existing CUDA kernel code to convert this to SYCL kernel code.
 
-This SYCL patterhn is able to re-use an existing CUDA kernel from a `*.cu` file into a C++ kernel, following these steps:
+A `*.cu` file is a combination of the Host CUDA API and device CUDA kernel code.
 
-1. For the device kernel, we will introduce a SYCL functor that will encapsulate 
+Existing CUDA kernel code can be transformed into SYCL code using the pattern described in this 
+document.
+
+1. For the device kernel code, we will introduce a SYCL functor that will encapsulate 
 all the CUDA code, enabling direct re-use of CUDA kernels without major 
-modifications whenever using simple kernel cases.
+modifications for simple kernel use cases.
 
-1. For the host code, the CUDA API is manually replaced with equivalent SYCL/C++ code. 
+1. For the host code, the CUDA API is manually replaced with the equivalent SYCL C++ code. 
 
 ## CUDA device kernel
 

utils/cuda_to_sycl/convertor/add.cpp renamed to utils/cuda_to_sycl/converter/add.cpp

@@ -155,13 +155,13 @@ struct real_accessor_t<T1, acc_t<T2>> {
 };
 
 template <typename conv_t>
-struct convertor {
+struct converter {
  using type = conv_t;
  static conv_t inline convert(conv_t c, cl::sycl::handler&) { return c; }
 };
 
 template <typename T>
-struct convertor<T*> {
+struct converter<T*> {
  using type = real_accessor_t<T, acc_t<uint8_t>>;
  static type inline convert(T* vir_ptr, cl::sycl::handler& h) {
  return type(get_global_pointer_mapper().get_access(vir_ptr, h));
@@ -242,10 +242,10 @@ class CudaCommandGroup<KernelT<Param_t...>> {
  utility::tuple::IndexList<Is...>) {
  auto globalrange = (gridSize_ * blockSize_);
  using u_ker_t = KernelT<dim3, dim3, dim3, dim3, cl::sycl::nd_item<3>,
- typename convertor<append_param_t>::type...>;
+ typename converter<append_param_t>::type...>;
  using kernel_t = kernel_dispatcher<u_ker_t>;
  auto func = kernel_t((
- convertor<append_param_t>::convert(utility::tuple::get<Is>(t2), h))...);
+ converter<append_param_t>::convert(utility::tuple::get<Is>(t2), h))...);
  h.parallel_for(
  nd_range<3>{
  cl::sycl::range<3>(globalrange.x, globalrange.y, globalrange.z),