feat(archive): keep page open on captcha-status errors so callers can promote

Adds OpenPageOptions.AllowNonOKStatus. When set, openPage no longer closes
the page on non-2xx (other than 404) and Open returns both a usable Document
and ErrInvalidStatusCode. archive.IsArchived and Archive opt in, so callers
can PromoteToInteractive the captcha page, hand it to a human solver, and
demote back to extract content from the same browser instance — avoiding
the cf_clearance fingerprint-binding issue that re-challenges any fresh
retry browser.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

sites/archive/archive.go

@@ -57,15 +57,25 @@ func (c Config) IsArchived(ctx context.Context, b extractor.Browser, target stri
 
 	slog.Info("checking if url is archived", "url", uri.String(), "config", c, "endpoint", endpoint)
 
-	doc, err := b.Open(ctx, uri.String(), extractor.OpenPageOptions{})
+	doc, err := b.Open(ctx, uri.String(), extractor.OpenPageOptions{AllowNonOKStatus: true})
 
 	if err != nil {
+		if errors.Is(err, extractor.ErrPageNotFound) {
+			if doc != nil {
+				_ = doc.Close()
+			}
+			return nil, nil
+		}
+		// On ErrInvalidStatusCode (e.g. archive.ph 403 with Cloudflare
+		// captcha) the page is kept open by AllowNonOKStatus so the caller
+		// can promote it to an InteractiveBrowser and let a human solve
+		// the challenge. Return both the doc and the wrapped error.
+		if errors.Is(err, extractor.ErrInvalidStatusCode) && doc != nil {
+			return doc, fmt.Errorf("failed to open url: %w", err)
+		}
 		if doc != nil {
 			_ = doc.Close()
 		}
-		if errors.Is(err, extractor.ErrPageNotFound) {
-			return nil, nil
-		}
 		return nil, fmt.Errorf("failed to open url: %w", err)
 	}
 
@@ -95,9 +105,15 @@ func (c Config) Archive(ctx context.Context, b extractor.Browser, target string)
 		return nil, fmt.Errorf("invalid endpoint: %w", err)
 	}
 
-	doc, err := b.Open(ctx, c.Endpoint, extractor.OpenPageOptions{})
+	doc, err := b.Open(ctx, c.Endpoint, extractor.OpenPageOptions{AllowNonOKStatus: true})
 
 	if err != nil {
+		// On ErrInvalidStatusCode (e.g. archive.ph 403 with Cloudflare
+		// captcha) the page is kept open by AllowNonOKStatus so the caller
+		// can promote it. Return both the doc and the wrapped error.
+		if errors.Is(err, extractor.ErrInvalidStatusCode) && doc != nil {
+			return doc, fmt.Errorf("failed to open url: %w", err)
+		}
 		if doc != nil {
 			_ = doc.Close()
 		}