Disallow breaking on rb tags when layout aware scanning is enabled

[Kuuuube]

rb tags should be ignored by layout aware scanning. This bug only appears on firefox presumably due to chromium not properly supporting rb tags.

Fixes #888

[jamesmaa]

Are there cases in which the rb content should not be ignored?

[Kuuuube]

This change does not make it ignore rb content. This makes it so it doesn't interpret rb tags as a line break and it instead keeps scanning further.

Here is an example of an rb tag being used:

<ruby><rb>美</rb><rt>お</rt><rb>味</rb><rt>い</rt></ruby>しい

Yomitan on firefox with layout aware scanning will only allow scanning 美, 味, or しい. not 美味しい. This change allows it to instead scan 美味しい.

For clarity, rb tags are deprecated and should not be used but they are still present in some cases. The correct way to represent the above word with furigana would be:

<ruby>美<rt>お</rt>味<rt>い</rt></ruby>しい

[jamesmaa]

Dumb, unrelated question but how is yomitan avoiding scanning 美お味いしい?

[Kuuuube]

Dumb, unrelated question but how is yomitan avoiding scanning 美お味いしい?

Not totally sure how it works but we have this which seems to be doing some sort of filtering:

https://github.com/themoeway/yomitan/blob/dd1f195cd721948732abbbb2428999393844487e/ext/js/dom/dom-text-scanner.js#L393-L410

If you try to scan an rt tag Yomitan just wont let you.

[Kuuuube]

Apparently not marking the tag as enterable slightly breaks the {sentence} handlebar even though scanning works fine. Fixed that. This should be totally good to go now.