[API] Propose to support LeakyRelu operation

[NALLEIN]

Computes rectified linear activation on the input tensor element-wise. This is a custom op that can be achieved with max and mul

• The output is calculated using this formula: output = max(mul(alpha, input),input)
• Inputs:
• • 0: A tensor, specifying the input.
• • 1: An INT 32 Value , specifying the scaling factor for negative values(provided alpha <= 1).
• Outputs:
• • 0: The output tensor of same shape as input0.

Model Format	API reference
Tensorflow.lite	https://www.tensorflow.org/lite/guide/ops_compatibility LEAKY_RELU
ONNX	https://github.com/onnx/onnx/blob/master/docs/Operators.md#LeakyRelu
OpenVINO	https://docs.openvinotoolkit.org/latest/_docs_MO_DG_prepare_model_convert_model_IR_V10_opset1.html#Maximum https://docs.openvinotoolkit.org/latest/_docs_MO_DG_prepare_model_convert_model_IR_V10_opset1.html#Multiply

backend	support
WASM	Yes
WebGL	Yes
NNAPI
MPS
BNNS
clDNN
MKLDNN
DirectML
IE-clDNN
IE-MKLDNN