feat(geschichten): add stripHtml util and GeschichtenCard component

stripHtml() strips tags via DOMParser (browser) with a regex fallback for
SSR. plainExcerpt() truncates at a word boundary with an ellipsis. Both
covered by Vitest specs.

GeschichtenCard renders the top 3 published stories about a person on
/persons/[id], with an editorial excerpt, publication date, author, and a
"+ Geschichte schreiben" link visible only to BLOG_WRITERs. Footer link to
/geschichten?personId=... appears once geschichten.length >= 3.

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

frontend/src/lib/utils/stripHtml.ts

@@ -0,0 +1,23 @@
+/**
+ * Strip HTML tags from a string and return the plain text.
+ * Uses DOMParser in the browser, falls back to a regex strip on the server
+ * (where DOMParser is not available without isomorphic-dompurify's JSDOM).
+ */
+export function stripHtml(html: string | null | undefined): string {
+	if (!html) return '';
+	if (typeof DOMParser === 'function') {
+		const doc = new DOMParser().parseFromString(html, 'text/html');
+		return (doc.body.textContent ?? '').trim();
+	}
+	return html.replace(/<[^>]*>/g, '').trim();
+}
+
+/**
+ * Strip HTML and truncate to a maximum length, appending an ellipsis when
+ * the source exceeds it. Used for editorial story excerpts.
+ */
+export function plainExcerpt(html: string | null | undefined, max = 80): string {
+	const text = stripHtml(html);
+	if (text.length <= max) return text;
+	return text.slice(0, max).replace(/\s+\S*$/, '') + '…';
+}