DNN: add the Winograd fp16 support (#23654)

* add Winograd FP16 implementation

* fixed dispatching of FP16 code paths in dnn; use dynamic dispatcher only when NEON_FP16 is enabled in the build and the feature is present in the host CPU at runtime

* fixed some warnings

* hopefully fixed winograd on x64 (and maybe other platforms)

---------

Co-authored-by: Vadim Pisarevsky <vadim.pisarevsky@gmail.com>

modules/dnn/include/opencv2/dnn/all_layers.hpp

@@ -303,7 +303,7 @@ CV__DNN_INLINE_NS_BEGIN
         // quantization type flag. The perChannel default is true, that means it contains the parameters
         // of per-Channel quantization. Otherwise, that means this layer contains per-Tensor quantized parameters.
         bool per_channel;
-        bool useWinograd = true; // Flag whether to use Winograd to speed up 3x3 convolution.
+        bool useWinograd = false; // Flag whether to use Winograd to speed up 3x3 convolution.
         static Ptr<BaseConvolutionLayer> create(const LayerParams& params);
     };
 

modules/dnn/include/opencv2/dnn/dnn.hpp

@@ -1458,6 +1458,9 @@ CV__DNN_INLINE_NS_BEGIN
          /// @sa Net::setPreferableTarget
          CV_WRAP Model& setPreferableTarget(dnn::Target targetId);
 
+         /// @sa Net::enableWinograd
+         CV_WRAP Model& enableWinograd(bool useWinograd);
+
          CV_DEPRECATED_EXTERNAL
          operator Net&() const { return getNetwork_(); }