fix(charset): handle quoted, spaced, and case-insensitive charset values

[vmphase]

The current CHARSET_RE only matches unquoted/unspaced/lowercase form of the charset= param, which does not match RFC 9110's grammar.

CHARSET_RE = re.compile(r"charset=([\w-]+)")

CHARSET_RE.search('text/html; charset="utf-8"')   # None
CHARSET_RE.search('text/html; charset = utf-8')   # None
CHARSET_RE.search('text/html; Charset=utf-8')     # None
CHARSET_RE.search('text/html; CHARSET="UTF-8"')   # None

# only this works fine
CHARSET_RE.search('text/html; charset=utf-8')     # <Match: 'utf-8'>  

Checklist

• I have manually reviewed the changes and fully understand the code.