Enhance page.rs to support table extraction and title extraction

[Rick9622092]

At present, the to_text_page method of page.rs can only obtain image and text information, and cannot obtain path information, which is required in table extraction; Suggest adding methods for traversing display objects within the page. In addition, TextChar does not encapsulate the method of obtaining font, which is required in applications such as inferring whether it is a title based on the font.