feat: add new config option: model_supports_reasoning_summaries (#1524)

As noted in the updated docs, this makes it so that you can set:

```toml
model_supports_reasoning_summaries = true
```

as a way of overriding the existing heuristic for when to set the
`reasoning` field on a sampling request:


https://github.com/openai/codex/blob/341c091c5b09dc706ab5c7d629516e6ef5aaf902/codex-rs/core/src/client_common.rs#L152-L166

codex-rs/exec/src/event_processor.rs

@@ -139,7 +139,7 @@ impl EventProcessor {
             ("sandbox", summarize_sandbox_policy(&config.sandbox_policy)),
         ];
         if config.model_provider.wire_api == WireApi::Responses
-            && model_supports_reasoning_summaries(&config.model)
+            && model_supports_reasoning_summaries(config)
         {
             entries.push((
                 "reasoning effort",