Wrong display of reasoning setting with gpt-5.1-codex and gpt-5.1

What version of Codex is running?

codex-cli 0.58.0

What subscription do you have?

ChatGPT Plus

Which model were you using?

gpt-5.1-codex

What platform is your computer?

Darwin 24.3.0 arm64 arm

What issue are you seeing?

Currently when using gpt-5.1 in $CODEX_HOME/config.toml without any model_reasoning_effort set i.e.:

```config.toml
model = "gpt-5.1"

and asking for `/status` after launching a session you get for the model part:

```shell
Model:            gpt-5.1 (reasoning none, summaries auto)

while it is clearly stated in codex-rs/core/src/model_family.rs that the default reasoning is medium:

    } else if slug.starts_with("gpt-5.1") {
        model_family!(
            slug, "gpt-5.1",
            supports_reasoning_summaries: true,
            apply_patch_tool_type: Some(ApplyPatchToolType::Freeform),
            support_verbosity: true,
            default_verbosity: Some(Verbosity::Low),
            base_instructions: GPT_5_1_INSTRUCTIONS.to_string(),
            default_reasoning_effort: Some(ReasoningEffort::Medium),
        )

The underlying issue is that currently there is no fallback to model_family for reasoning effort display. The same applies to gpt-5.1-codex where the default API behaviour is probably medium but in this case it also show reasoning none

What steps can reproduce the bug?

Change config.toml to:

```config.toml
model = "gpt-5.1"

or run `codex -m gpt-5.1` without any `model_reasoning_effort` set in `config.toml` and then `/status`

### What is the expected behavior?

`/status` for the above case should yield:

```shell
Model:            gpt-5.1 (reasoning medium, summaries auto)

Additional information

_No response_