IVGCVSW-5815 Generalise ConstCpuTensorHandle

* Generalises ConstCpuTensorHandle and inherited
  classes by removing 'Cpu' from aliases.
* New renamed classes: ConstTensorHandle, TensorHandle,
  ScopedTensorHandle, PassthroughTensorHandle,
  ConstPassthroughTensorHandle.

Signed-off-by: James Conroy <james.conroy@arm.com>
Change-Id: I1824e0e134202735fb77051f20a7252f161dfe16
diff --git a/src/backends/backendsCommon/test/DynamicBackendTests.hpp b/src/backends/backendsCommon/test/DynamicBackendTests.hpp
index 8302bfd..a4f1613 100644
--- a/src/backends/backendsCommon/test/DynamicBackendTests.hpp
+++ b/src/backends/backendsCommon/test/DynamicBackendTests.hpp
@@ -9,8 +9,8 @@
 #include <armnn/backends/DynamicBackend.hpp>
 #include <armnn/backends/ILayerSupport.hpp>
 #include <armnn/utility/PolymorphicDowncast.hpp>
-#include <backendsCommon/CpuTensorHandle.hpp>
 #include <backendsCommon/DynamicBackendUtils.hpp>
+#include <backendsCommon/TensorHandle.hpp>
 #include <Filesystem.hpp>
 #include <reference/workloads/RefConvolution2dWorkload.hpp>
 #include <Runtime.hpp>
@@ -1473,7 +1473,7 @@
         { outputInfo }
     };
     convolution2dQueueDescriptor.m_Inputs.push_back(nullptr);
-    auto weights = std::make_unique<ScopedCpuTensorHandle>(weightInfo);
+    auto weights = std::make_unique<ScopedTensorHandle>(weightInfo);
     convolution2dQueueDescriptor.m_Weight = weights.get();
 
     // Create a convolution workload with the dummy settings