feat: Nvidia API direkt-Anbindung mit Thinking-Mode Support

Neues Server-Setting NvidiaApiKey – wenn gesetzt, wird direkt
integrate.api.nvidia.com verwendet statt g4f.space Proxy.
Thinking-Modelle (kimi-k2, qwq, deepseek-r1) erhalten automatisch
chat_template_kwargs und reasoning-Output wird als klappbare
Denkprozess-Bubble im Chat angezeigt.

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

Services/ServerSettingsService.cs

@@ -36,6 +36,7 @@ public class ServerSettingsService
         new SettingDefinition( "ScanDailyHour",      "int",    "-1",    "Key-Scanner Full-Scan Uhrzeit (0-23, -1 = deaktiviert)",         "number"   ),
         new SettingDefinition( "GitHubEventScanner", "bool",   "false", "GitHub Event Scanner (Echtzeit-Monitoring öffentlicher Pushes)", "toggle"   ),
         new SettingDefinition( "HuggingFaceToken",   "string", "",      "HuggingFace Token (optional, erhöht Rate-Limits)",            "password" ),
+        new SettingDefinition( "NvidiaApiKey",       "string", "",      "Nvidia API-Key (direkt statt g4f.space Proxy)",                "password" ),
         new SettingDefinition( "ScannerKeyTypes",    "string_array", "anthropic_oauth,krea_ai", "Key-Scanner: Aktive Key-Typen",                               "checkboxes" ),
     };