feat: Dynamisches Modell-Laden für Nvidia & Pollinations Provider

- OpenAiService: FetchModels() generisch für beide Provider (OpenAI-Format & Pollinations-Array-Format)
- Pollinations Base-URL direkt auf text.pollinations.ai/openai (kein g4f.space Proxy → kein 402)
- Pollinations Modelle-URL: text.pollinations.ai/models
- ModelsEndpoint: /api/models/nvidia + /api/models/pollinations
- agent-page.js: loadProviderModels() generisch, Promise.all für beide Provider parallel
- index.html: statische Nvidia-Option entfernt, beide Optgroups mit 'Wird geladen...'

wwwroot/index.html

@@ -191,8 +191,11 @@
           <option value="opus-4-5">Opus 4.5</option>
           <option value="opus-4-6" selected>Opus 4.6</option>
           <option value="haiku-4-5">Haiku 4.5</option>
-          <optgroup label="Nvidia (via g4f.space)">
-            <option value="nvidia:moonshotai/kimi-k2.5">Kimi K2.5 (Moonshot)</option>
+          <optgroup label="Pollinations (via g4f.space)" id="pollinationsOptgroup">
+            <option value="" disabled>Wird geladen...</option>
+          </optgroup>
+          <optgroup label="Nvidia (via g4f.space)" id="nvidiaOptgroup">
+            <option value="" disabled>Wird geladen...</option>
           </optgroup>
         </select>
         <div class="more">

wwwroot/js/pages/agent-page.js

@@ -18,6 +18,32 @@ registerPage('agent', {
     const nameEl = document.getElementById('agentName');
     const descEl = document.getElementById('agentDescription');
     const modelEl = document.getElementById('agentModel');
+    const nvidiaOptgroup       = document.getElementById('nvidiaOptgroup');
+    const pollinationsOptgroup = document.getElementById('pollinationsOptgroup');
+
+    // Modelle dynamisch laden
+    const loadProviderModels = async (url, prefix, optgroup) => {
+      try {
+        const resp = await fetch(url);
+        const data = await resp.json();
+        if (optgroup && data.models && data.models.length > 0) {
+          optgroup.innerHTML = '';
+          for (const m of data.models) {
+            const opt = document.createElement('option');
+            opt.value = prefix + ':' + m.id;
+            opt.textContent = m.displayName;
+            optgroup.appendChild(opt);
+          }
+        }
+      } catch (e) {
+        if (optgroup) optgroup.innerHTML = '<option value="" disabled>Fehler beim Laden</option>';
+      }
+    };
+
+    await Promise.all([
+      loadProviderModels('/api/models/nvidia',       'nvidia',       nvidiaOptgroup),
+      loadProviderModels('/api/models/pollinations', 'pollinations', pollinationsOptgroup)
+    ]);
     const allowUserChoiceEl = document.getElementById('agentAllowUserChoice');
     const autoCommitEl = document.getElementById('agentAutoCommit');
     const useSoulEl = document.getElementById('agentUseSoul');