Automatically detect code language in reading mode for better highlighting

Clear and concise description of the problem

In reading mode, mozilla/readability causes the code highlighting effect and language annotation attributes to disappear, but I noticed that shiki still highlights the content. It would be better if the language could be automatically detected before highlighting.

Suggested solution

In the past, I ported guesslang, a library that detects the programming language of source code through deep learning, to run in the browser using TensorFlow.js.
I added the esm format and web worker calls to it and used it in a production environment. You can also refer to this online demo:

https://ray-d-song.github.io/guesslang-js/

Alternative

No response

Additional context

No response

Validations

• Check that there isn't already an issue that request the same feature to avoid creating a duplicate.
• This issue is valid

Trace how reading mode uses mozilla/readability and how shiki receives code before highlighting. Review the referenced guesslang-js demo and its browser, ESM, and web-worker approach, then define how automatic language detection should restore highlighting and language attributes in reading mode.