IVGCVSW-5005 Fix Deprecated Functions in CL/NEON Resize Workload

 * Update CL pin to f3ad9513dd46fca1d6c5e4550286480fdbaba056

IVGCVSW-5005: Fix Deprecated Functions in CL/NEON Resize Workload
 * Add missing virtual function to NeonIntercepterScheduler
 * Update CL/Neon Resize Workloads to use ScaleKernelInfo
 * Update CL/Neon Resize workloads to set correct Sampling Policy for Half Pixels

IVGCVSW-4981: Change CL/NEON Softmax axis value
 * Default value is now 0

Signed-off-by: David Monahan <david.monahan@arm.com>
Change-Id: I8a14c6a64e5a24bc1a66a7d3685cd388c2245702
diff --git a/src/backends/aclCommon/ArmComputeUtils.hpp b/src/backends/aclCommon/ArmComputeUtils.hpp
index 0ee13b3..eae152d 100644
--- a/src/backends/aclCommon/ArmComputeUtils.hpp
+++ b/src/backends/aclCommon/ArmComputeUtils.hpp
@@ -153,18 +153,21 @@
 template<typename T>
 inline T ComputeSoftmaxAclAxis(const SoftmaxDescriptor& softmaxDesc, const armnn::TensorInfo& tensor)
 {
-    // Detect the Android default value of -1 and return the ACL default value of 1.
+    // Detect the Android default value of -1 and return the ACL default value of 0.
     if (softmaxDesc.m_Axis == -1)
     {
-        return 1;
+        return 0;
     }
 
-   unsigned int dim = tensor.GetNumDimensions();
+    unsigned int dim = tensor.GetNumDimensions();
 
     ARMNN_ASSERT(dim != 0);
 
     // Currently ArmNN support axis 1.
-    return static_cast<T>(dim) - 1;
+    auto aclAxis = (static_cast<T>(dim) - 1);
+    aclAxis = aclAxis > 0 ? aclAxis -1 : aclAxis;
+
+    return aclAxis;
 }
 
 inline std::set<unsigned int> ComputeSplitAxis(const armnn::SplitterDescriptor& desc, const TensorShape& input)