Add mathjax nodes to extracted content

[kiro-liu]

I want to extract MathJax syntax from the html page,
But In the process of using it I found that the mathematical formulas in the page would be discarded
just like this:

PAGE: "https://opg.optica.org/oe/fulltext.cfm?uri=oe-19-10-9848&id=213654"

[adbar]

@Ryannees Yes, this tag is not covered (yet).

[kiro-liu]

@adbar I used this html above to optimize this problem
Diff Result：

At the same time, I also tested some other sample parsing with mathml syntax and the results were greatly improved.
Maybe one PR will be submitted later, use stepai's name