Performance Optimization of Deep Learning Sparse Matrix Kernels on Intel Max Series GPU