IVGCVSW-5685 Add CpuAcc specific configuration option numberOfThreads

 * Added ability to set number of threads used in CpuAcc backend
 * Enabled number-of-threads option in ExecuteNetwork
 * Added TfLiteDelegate ModelOptions test
 * Added unsigned int type to BackendOptions.hpp

Signed-off-by: Matthew Sloyan <matthew.sloyan@arm.com>
Change-Id: Ia576d4f45cbe5df3654bc730bb5ebd5181d82b5a
diff --git a/tests/InferenceModel.hpp b/tests/InferenceModel.hpp
index 936d0bf..d20bb22 100644
--- a/tests/InferenceModel.hpp
+++ b/tests/InferenceModel.hpp
@@ -99,6 +99,7 @@
     bool                            m_EnableFastMath;
     bool                            m_SaveCachedNetwork;
     std::string                     m_CachedNetworkFilePath;
+    unsigned int                    m_NumberOfThreads;
 
     Params()
         : m_ComputeDevices{}
@@ -113,6 +114,7 @@
         , m_EnableFastMath(false)
         , m_SaveCachedNetwork(false)
         , m_CachedNetworkFilePath("")
+        , m_NumberOfThreads(0)
     {}
 };
 
@@ -436,7 +438,8 @@
             });
             armnn::BackendOptions cpuAcc("CpuAcc",
             {
-                { "FastMathEnabled", params.m_EnableFastMath }
+                { "FastMathEnabled", params.m_EnableFastMath },
+                { "NumberOfThreads", params.m_NumberOfThreads }
             });
             options.m_ModelOptions.push_back(gpuAcc);
             options.m_ModelOptions.push_back(cpuAcc);