On the realism of the rain microphysics representation of a squall line in the WRF model. Part I: Evaluation with multifrequency cloud radar doppler spectra observations