binshe

### Description of the bug ``` >>> import pymupdf >>> pymupdf.version ('1.24.4', '1.24.2', '20240516000001') >>> doc = pymupdf.open("test3.pdf") >>> doc._get_page_labels() [(0, ''), (13, ''), (15, '')] >>> doc[0].get_label() 'B' ```...

feat: transform - ignore file fetching for html, and add item content fetching for json

5

- html: no need to fetch item links to files like .pdf, so just add a param to specify the fileExt to fetch (default is html, htm, and no ext),...