Improve parsing in readBody

[itpropro]

Describe the feature

I would propose the following optimizations in readBody and would like to know what you think about that @pi0

• Instead of first calling response.text(), use the content-length header to determine if a body is existent (Transfer-Encoding: chunked and form data shouldn't be relevant here, as they have their own helpers)
• Then check the content-type header and call parseURLEncodedBody on request.text() if applies
• Otherwise directly return response.json() instead of passing text to JSON.parse as destr was removed anyway

This would avoid memory overhead for large files and could also enable file/request size limits, which are normally based on the content-length header (DOS protection). Although there might be better places for content-length validation.

Additional information

• Would you be willing to help implement this feature?