Assume thаt tensоrflоw hаs аlready been impоrted as tf. Write a Python function named create_cnn_model() that returns a convolutional neural network for binary classification with the following architecture: An input layer for images with shape (28, 28, 1). A convolutional layer with 32 filters, a 3 × 3 kernel, and tanh activation. A max pooling layer with pool size 2 × 2. A flattening layer. A fully connected hidden layer with 16 neurons and ReLU activation. An output layer with 1 neuron and sigmoid activation.