Text this: ETIA: Enhancing Text2Image Surround View Scene Generation With Semantic Annotation via Diffusion for Autonomous Driving