feat: Dashboard-Metriken nach Mailpiler-Vorbild (Uptime, Aktivität, Prognose)

- storage_stats.go (neu): MailActivityStats (60min/24h/7d/30d), StorageEstimateStats
- dashboard_handlers.go: Uptime (/proc/uptime), activity + estimate in System-Stats-Response
- DashboardTab: Uptime in API-Kachel, neue Kacheln "Mail-Eingang" + "Speicherprognose"
- Warnung (Badge "Knapp!") wenn Partition in <90 Tagen voll

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

src/lib/api/system.ts

@@ -83,14 +83,31 @@ export interface SystemStatsMailInfo {
   subject: string;
 }
 
+export interface SystemStatsActivity {
+  last_60_min: number;
+  last_24h: number;
+  last_7d: number;
+  last_30d: number;
+}
+
+export interface SystemStatsEstimate {
+  avg_mails_per_day: number;
+  avg_mail_bytes: number;
+  days_until_full: number; // -1 = unknown
+  archive_age_days: number;
+}
+
 export interface SystemStats {
   cpu: SystemStatsCPU;
   ram: SystemStatsRAM;
   disks: SystemStatsDisk[];
+  uptime: { seconds: number };
   archive: {
     first_mail: SystemStatsMailInfo | null;
     last_mail: SystemStatsMailInfo | null;
   };
+  activity: SystemStatsActivity;
+  estimate: SystemStatsEstimate;
 }
 
 export interface SecurityCheck {