Polish month names are not recognized (date reduced to the year) #206

htmldate 1.11.0 (dateparser 1.4.3, Python 3.13)

Polish dates written with a month name are not extracted when they are part of a longer string: extensive mode returns only the year, fast mode returns nothing.

from htmldate import find_date

doc = '<html><body><p class="date">Opublikowano: 12 października 2025</p></body></html>'
find_date(doc)                          # '2025-01-01', expected '2025-10-12'
find_date(doc, extensive_search=False)  # None, expected '2025-10-12'

The German and English equivalents (Veröffentlicht: 12. Oktober 2025, Published: 12 October 2025) return 2025-10-12 in both modes. The same happens through trafilatura 2.3.1 (extract_metadata(...).date == '2025-01-01').

Cause: regex_parse only knows the names in MONTHS (EN/FR/DE/ID/TR), so Polish dates depend on dateparser. With STRICT_PARSING, dateparser rejects the string because of the prefix ("Opublikowano:" means "Published:"), and the year-only fallback wins. Fast mode never calls dateparser.

A PR adding the Polish month names to MONTHS follows.