feat: add GPT-5-Codex support with Responses API integration

Adds support for OpenAI's GPT-5-Codex model which uses the new Responses API
endpoint (/v1/responses) instead of the standard Chat Completions API.

Changes:
- Add GPT-5-Codex to MODEL_CAPABILITIES with 400K context, 128K output
- Prioritize GPT-5-Codex for EXTENDED_REASONING tasks
- Add aliases: codex, gpt5-codex, gpt-5-code
- Update tests to expect GPT-5-Codex for extended reasoning

Benefits:
- 40-80% cost savings through Responses API caching
- 3% better performance on coding tasks (SWE-bench)
- Leverages existing dual-API infrastructure

tests/test_auto_mode_provider_selection.py

@@ -98,7 +98,7 @@ class TestAutoModeProviderSelection:
             balanced = ModelProviderRegistry.get_preferred_fallback_model(ToolModelCategory.BALANCED)
 
             # Should select appropriate OpenAI models based on new preference order
-            assert extended_reasoning == "o3"  # O3 for extended reasoning
+            assert extended_reasoning == "gpt-5-codex"  # GPT-5-Codex prioritized for extended reasoning
             assert fast_response == "gpt-5"  # gpt-5 comes first in fast response preference
             assert balanced == "gpt-5"  # gpt-5 for balanced