Thai word segmentation (tokenization): split Thai text, which has no spaces between words, into a list of words. Body: {"text": "..."} (max 5000 chars).