feat(codex): add xAI (Grok) native Responses preset

API-key preset pointing at https://api.x.ai/v1 with wire_api =
"responses": xAI serves the Responses API natively, so no route
takeover conversion is needed. Ships a grok-4.5 catalog entry
(500K context, parallel tools, text+image) so the model shows up
under Codex's strict catalog parsing. This is deliberately not a
managed-account preset: providerType stays absent, pinned by a
regression test alongside the Claude-side xAI OAuth presets.

src/config/codexProviderPresets.ts

@@ -999,6 +999,30 @@ requires_openai_auth = true`,
     icon: "novita",
     iconColor: "#000000",
   },
+  {
+    name: "xAI (Grok)",
+    websiteUrl: "https://x.ai/api",
+    apiKeyUrl: "https://console.x.ai",
+    auth: generateThirdPartyAuth(""),
+    config: generateThirdPartyConfig("xai", "https://api.x.ai/v1", "grok-4.5"),
+    endpointCandidates: ["https://api.x.ai/v1"],
+    // xAI 官方以 /v1/responses 为一等端点（docs.x.ai api-reference）：Codex 硬依赖的
+    // store:false / include=["reasoning.encrypted_content"] / reasoning effort 均支持，
+    // 原生 Responses，无需路由接管转换
+    apiFormat: "openai_responses",
+    modelCatalog: modelCatalog([
+      {
+        model: "grok-4.5",
+        displayName: "Grok 4.5",
+        contextWindow: 500000,
+        supportsParallelToolCalls: true,
+        inputModalities: ["text", "image"],
+      },
+    ]),
+    category: "third_party",
+    icon: "xai",
+    iconColor: "#000000",
+  },
   {
     name: "Nvidia",
     websiteUrl: "https://build.nvidia.com",