fix(automod): flow real LLM analysis + descriptive fallback

Root cause: ai-analysis-worker read llmResult.explanation and
llmResult.toxicityScore — fields the LLM pipeline never produces
(canonical AnalysisResult uses analysis/score). Every message fell back
to the bare template "Tidak ada indikasi pelanggaran." and the stored
score was always 0.

- Map analysis/score correctly; fallback now quotes the message content
- Prompt: ban generic analysis phrasing, require reply context
- LLM context: include replied-to message content (metadata.reference)
  so the model can explain what the user is replying to
- Frontend: show thread/channel names from metadata instead of raw IDs
  (message card, detail views, search overlay); detail panel now
  displays the ai_analysis text
- Auto-delete log/DM include the descriptive analysis as the reason

services/frontend/src/components/messages/ai-analysis-panel.tsx

@@ -11,6 +11,7 @@ interface AiAnalysisPanelProps {
   categories?: string[] | string | null;
   action?: string | null;
   score?: number | null;
+  analysis?: string | null;
 }
 
 const severityColor: Record<string, string> = {
@@ -29,6 +30,7 @@ export function AiAnalysisPanel({
   categories,
   action,
   score,
+  analysis,
 }: AiAnalysisPanelProps) {
   if (!status || status === "pending") {
     return (
@@ -93,6 +95,12 @@ export function AiAnalysisPanel({
         </div>
       )}
 
+      {analysis && (
+        <p className="text-xs leading-relaxed text-text-secondary/90 border-l-2 border-glass-border pl-2">
+          {analysis}
+        </p>
+      )}
+
       {action && action !== "none" && (
         <div className="text-xs">
           <span className="text-text-secondary/60">Recommended: </span>