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...'

Endpoints/ModelsEndpoint.cs

@@ -0,0 +1,32 @@
+using System.Linq;
+using Gott.Services;
+using Microsoft.AspNetCore.Builder;
+using Microsoft.AspNetCore.Http;
+using Microsoft.AspNetCore.Routing;
+
+namespace Gott.Endpoints;
+
+public static class ModelsEndpoint
+{
+    public static void MapModelsEndpoint(this IEndpointRouteBuilder app)
+    {
+        app.MapGet("/api/models/nvidia",       HandleNvidiaModels);
+        app.MapGet("/api/models/pollinations", HandlePollinationsModels);
+    }
+
+    private static IResult HandleNvidiaModels(OpenAiService openAiService)
+    {
+        var models = openAiService.GetNvidiaModels()
+            .Select(m => new { id = m.Id, displayName = m.DisplayName })
+            .ToList();
+        return Results.Json(new { models });
+    }
+
+    private static IResult HandlePollinationsModels(OpenAiService openAiService)
+    {
+        var models = openAiService.GetPollinationsModels()
+            .Select(m => new { id = m.Id, displayName = m.DisplayName })
+            .ToList();
+        return Results.Json(new { models });
+    }
+}