Add archive scraping, webcal feed, and scrape schedule reporting

- scraper.py: --archive flag to scrape https://fundbureau.de/archiv.html
  (past events), fixing a container-specific wait selector and a crash
  on events with no door time that this surfaced.
- api: EventStore now merges an optional archive JSON file into the
  main event list (deduped by date+name); adds GET /events/archive.
- api: GET /calendar.ics serves an RFC 5545 feed of all events for
  webcal subscriptions, with an all-day fallback when no door time is
  parseable.
- api: GET /health reports last-scrape time (from file mtime) and, via
  new SCRAPE_CRON/ARCHIVE_SCRAPE_CRON env vars, the next scheduled run
  and seconds until it, for both the regular and archive scrape.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

scraper/scraper.py

@@ -8,6 +8,8 @@ load_local_file = False
 output_file = "fundi-scraped-output.json"
 output_path = "./"
 ignore_ticket_link = False
+ARCHIVE_URL = "https://fundbureau.de/archiv.html"
+ARCHIVE_OUTPUT_FILE = "fundi-archive-output.json"
 
 argp = argparse.ArgumentParser()
 argp.add_argument("--url", type=str, help="URL to scrape (default: https://fundbureau.de/)")
@@ -15,8 +17,12 @@ argp.add_argument("--output-file", type=str, help="Output JSON file name (defaul
 argp.add_argument("--output-path", type=str, help="Output path for the JSON file (default: current directory)")
 argp.add_argument("--load-local-file", action="store_true", help="Load local HTML file instead of scraping the website (default: False)")
 argp.add_argument("--ignore-ticket-link", action="store_true", help="Ignore ticket link extraction (will leave it empty in the output JSON)")
+argp.add_argument("--archive", action="store_true", help=f"Scrape the past-events archive ({ARCHIVE_URL}) instead of upcoming events. Changes the default url/output-file, both can still be overridden with --url/--output-file.")
 args = argp.parse_args()
 
+if vars(args)["archive"]:
+    scrape_url = ARCHIVE_URL
+    output_file = ARCHIVE_OUTPUT_FILE
 if vars(args)["url"]:
     scrape_url = vars(args)["url"]
 if vars(args)["output_file"]:
@@ -42,8 +48,13 @@ def scrape_website(url, load_local_file):
             page = browser.new_page()
             page.goto(url, wait_until="networkidle")
             try:
-                # Wait until at least one .event div actually renders in the DOM
-                page.wait_for_selector("#upcoming-events-container .event", timeout=15000)
+                # Wait until at least one .event div actually renders in the DOM.
+                # Upcoming events live in #upcoming-events-container, past/archive
+                # events in #past-events-container - match either.
+                page.wait_for_selector(
+                    "#upcoming-events-container .event, #past-events-container .event",
+                    timeout=15000,
+                )
             except Exception:
                 print("Warning: no .event elements appeared before timeout.")
             html = page.content()
@@ -86,7 +97,8 @@ def extract_events(soup, ignore_ticket_link):
                 })
 
 
-        event_starttime = event.find("div", class_="event-starttime").text.strip()
+        event_starttime_tag = event.find("div", class_="event-starttime")
+        event_starttime = event_starttime_tag.text.strip() if event_starttime_tag else ""
         if ignore_ticket_link:
             event_ticket_link = ""
             event_free = None