Infrastructure: Fix report.js unsupported buffer

[howard-e]

Relevant to address an issue found when running the npm run regression-report command

#959 (comment)

$ npm run regression-report

> [email protected] regression-report
> node test/util/report

Z:\aria-practices\node_modules\htmlparser2\lib\Tokenizer.js:628
            var c = this.buffer.charCodeAt(this._index);
                                ^

TypeError: this.buffer.charCodeAt is not a function
    at Tokenizer.parse (Z:\aria-practices\node_modules\htmlparser2\lib\Tokenizer.js:628:33)
    at Tokenizer.write (Z:\aria-practices\node_modules\htmlparser2\lib\Tokenizer.js:111:14)
    at Tokenizer.end (Z:\aria-practices\node_modules\htmlparser2\lib\Tokenizer.js:117:18)
    at Parser.end (Z:\aria-practices\node_modules\htmlparser2\lib\Parser.js:390:24)
    at parseDocument (Z:\aria-practices\node_modules\htmlparser2\lib\index.js:43:43)
    at Object.parseDOM (Z:\aria-practices\node_modules\htmlparser2\lib\index.js:58:12)
    at processDocumentationInExampleFiles (Z:\aria-practices\test\util\report.js:112:29)
    at Object.<anonymous> (Z:\aria-practices\test\util\report.js:214:1)
    at Module._compile (node:internal/modules/cjs/loader:1101:14)
    at Object.Module._extensions..js (node:internal/modules/cjs/loader:1153:10)

• htmlparser2.parseDOM was never meant to take a buffer. Relevant comment: Regression in #914 fb55/htmlparser2#991 (comment).
• htmlparser2.parseDOM is deprecated. Now using htmlparser2.parseDocument.