Zusammenfassung: keine '(Namen: ...)'-Zeile mehr unter jedem Absatz

Extraktion bleibt als Arbeitsdatum fuer den Rolling Context zwischen den Chunks.
Bestehende Summaries bereinigt (#74 7503 -> 7203 Zeichen).

app/llm.js

@@ -634,8 +634,12 @@ async function analyzeLong(raw, hint = CORRECTION_HINT) {
 
   // Deterministic conglomerate: chronological, one block per section with its
   // LLM headline. No LLM re-weighting = no mis-weighting. GUI renders pre-line.
+  // NOTE: the extracted proper nouns (s.names) are NOT printed under each block
+  // any more — they are working data for the rolling context between chunks
+  // (keeps the model's spelling of people/companies stable), not something the
+  // reader wants under every paragraph.
   const body = sections.map((s, k) =>
-    `${s.thema || `Teil ${k + 1}`}:\n${s.sum}${s.names.length ? `\n(Namen: ${s.names.join(', ')})` : ''}`
+    `${s.thema || `Teil ${k + 1}`}:\n${s.sum}`
   ).join('\n\n');
   const allNames = uniqStr(sections.flatMap(s => s.names));
   const summary = [overview, body, allNames.length ? `Genannte Namen: ${allNames.join(', ')}` : '']