[FEATURE] Subsetting based on textfiles

[the-hotmann]

I want to subset all fonts on my page. But not with basic standard subsets but instead custom ones.

So it basically should remove ALL chars that are not needed on my pages.
I would like to pass something like this to the config:

options: {
	subsetFromText: ['public/*.html', 'public/**/*.html'],    // can be TXT, HTML, all type of Textfile and should just respect the Text in the HTML so most probably no "<" and ">" if not in the Text
	formats: ['woff2']
},

I just want to have the WOFF2 and I would like to remove ALL unused chars which are not getting used on my pages.

A similar function is available grunt-ziti but it is archived and not supported anymore.

Is such subset option against HTML possible?