Text this: Enhanced CLIP-GPT Framework for Cross-Lingual Remote Sensing Image Captioning