refactor: restrict model checks to gpt-* for OpenAI/Codex models

- Updated model resolution logic to only check for gpt-* models, removing references to unsupported o1/o3 models in both model-resolver and provider-factory files.
- Enhanced comments for clarity regarding model support in Codex CLI.

apps/server/src/lib/model-resolver.ts

@@ -42,7 +42,8 @@ export function resolveModelString(
   }
 
   // OpenAI/Codex models - pass through unchanged
-  if (modelKey.startsWith("gpt-") || modelKey.startsWith("o")) {
+  // Only check for gpt-* models (Codex CLI doesn't support o1/o3)
+  if (modelKey.startsWith("gpt-")) {
     console.log(`[ModelResolver] Using OpenAI/Codex model: ${modelKey}`);
     return modelKey;
   }

apps/server/src/providers/provider-factory.ts

@@ -21,8 +21,9 @@ export class ProviderFactory {
   static getProviderForModel(modelId: string): BaseProvider {
     const lowerModel = modelId.toLowerCase();
 
-    // OpenAI/Codex models (gpt-*, o1, o3, etc.)
-    if (lowerModel.startsWith("gpt-") || lowerModel.startsWith("o")) {
+    // OpenAI/Codex models (gpt-*)
+    // Note: o1/o3 models are not supported by Codex CLI
+    if (lowerModel.startsWith("gpt-")) {
       return new CodexProvider();
     }