fix: set LLM_MODEL for Option 1 (cloud endpoints with NVIDIA API key)

- When Option 1 is selected, set LLM_MODEL to nvidia/llama-3.3-nemotron-super-49b-v1
- Previously only Option 2 (Brev) set LLM_MODEL
- Now both options correctly configure the LLM model
- Update summary output to show LLM_MODEL is set for Option 1

Fixes: LLM_MODEL not being set when using cloud endpoints with NVIDIA API key

Author: T-DevH

notebooks/setup/complete_setup_guide.ipynb

@@ -1469,9 +1469,29 @@
     "            # Remove EMBEDDING_API_KEY line if using Option 1 (will use NVIDIA_API_KEY)\n",
     "            content = re.sub(r'^EMBEDDING_API_KEY=.*$\\n?', '', content, flags=re.MULTILINE)\n",
     "        \n",
-    "        # Update LLM_MODEL if Brev model is provided (Option 2)\n",
-    "        if brev_model:\n",
-    "            # Check if LLM_MODEL exists in content, if not add it\n",
+    "        # Update LLM_MODEL\n",
+    "        # Option 1: Use default cloud model (nvidia/llama-3.3-nemotron-super-49b-v1)\n",
+    "        # Option 2: Use Brev model name if provided\n",
+    "        if choice == \"1\":\n",
+    "            # Option 1: Set to default cloud model\n",
+    "            default_cloud_model = \"nvidia/llama-3.3-nemotron-super-49b-v1\"\n",
+    "            if re.search(r'^LLM_MODEL=.*$', content, flags=re.MULTILINE):\n",
+    "                content = re.sub(\n",
+    "                    r'^LLM_MODEL=.*$',\n",
+    "                    f'LLM_MODEL={default_cloud_model}',\n",
+    "                    content,\n",
+    "                    flags=re.MULTILINE\n",
+    "                )\n",
+    "            else:\n",
+    "                # Add LLM_MODEL after NVIDIA_API_KEY if it doesn't exist\n",
+    "                content = re.sub(\n",
+    "                    r'^(NVIDIA_API_KEY=.*)$',\n",
+    "                    rf'\\1\\nLLM_MODEL={default_cloud_model}',\n",
+    "                    content,\n",
+    "                    flags=re.MULTILINE\n",
+    "                )\n",
+    "        elif brev_model:\n",
+    "            # Option 2: Use Brev model name if provided\n",
     "            if re.search(r'^LLM_MODEL=.*$', content, flags=re.MULTILINE):\n",
     "                content = re.sub(\n",
     "                    r'^LLM_MODEL=.*$',\n",
@@ -1572,6 +1592,7 @@
     "        print(\"=\" * 60)\n",
     "        if choice == \"1\":\n",
     "            print(\"   • NVIDIA_API_KEY: Set (will be used for all services)\")\n",
+    "            print(\"   • LLM_MODEL: Set (nvidia/llama-3.3-nemotron-super-49b-v1)\")\n",
     "        else:\n",
     "            print(\"   • NVIDIA_API_KEY: Set (Brev API key for LLM)\")\n",
     "            print(\"   • EMBEDDING_API_KEY: Set (NVIDIA API key for Embedding)\")\n",