I have some Questions about the LanguageLevel of the parser.

1. What is difference between CURRENT and BLEEDING_EDGE? The descriptions in the Javadoc are slightly different, but I cannot capture what is the different between "latest Java version" and "newest Java features". (Lately only BLEEDING_EDGE was updated by @johannescoetzee to 21, but CURRENT is still 16.)
2. The RAW option disable post processing and validation. I understand that validation means that the code is checked to be valid for the specific Java version (e.g., Text Block Literals are not supported by Java versions lower than 15). If I want to parse all Java versions, I do not care about such validation, but what does the "post processing" do?
3. The supported Java versions are also mentioned in the README (1.0 to 18) and in the "About" section of the GitHub Repository (1 to 17). Should these versions only correspond to latest release or the status of the main branch? The range from the README actual correspond to the latest release (3.25.10).