Fix SRY (Surrey County Council) scraper

[symroe]

What broke

Surrey's ModGov endpoint (mycouncil.surreycc.gov.uk) is behind Incapsula WAF protection. Plain wreq requests return a 212-byte Incapsula JS-challenge HTML stub instead of the councillor XML, resulting in 0 councillors scraped (the scraper ran without error but got no data). Additionally, the base_url in metadata.json had a trailing slash that produced a double-slash in the constructed API URL (https://mycouncil.surreycc.gov.uk//mgWebService.asmx/GetCouncillorsByWard).

What was fixed

• Added http_lib = "playwright" to the Scraper class — headless Chromium executes the Incapsula JS challenge, sets the required cookies, and triggers a redirect to the actual XML endpoint. Chrome's XML viewer embeds the raw XML elements in the page DOM, so BeautifulSoup's xml parser successfully finds all ward/councillor elements.
• Removed trailing slash from base_url in metadata.json

Scrape results

Metric	Count
Councillors found	81
With email address	79
With photo	81

Note: Local verification used ignore_https_errors=True in playwright because the locally-downloaded chromium-headless-shell has a restricted CA bundle that does not trust mycouncil.surreycc.gov.uk's certificate. The Lambda container image ships with a system-linked Chromium where this cert is trusted, matching the behaviour seen in COT (PR #334). The councillor data above was confirmed via direct playwright scraping with cert errors bypassed.

---

Generated by Claude Code

[symroe]

Re-scrape after f4f1945

Added http_lib = "playwright" and removed trailing slash in base_url.

Metric	Count
Councillors found	81
With email address	79
With photo	81

Verified via playwright with ignore_https_errors=True to simulate Lambda's trusted cert environment. The Incapsula JS challenge completes in headless Chrome, and Chrome's XML viewer DOM embeds the actual XML elements that BeautifulSoup's xml parser successfully finds.

---

Generated by Claude Code