Preserve image detail in app-server inputs (#20693)

## Summary

- Add optional image detail to user image inputs across core, app-server
v2, thread history/event mapping, and the generated app-server
schemas/types.
- Preserve requested detail when serializing Responses image inputs:
omitted detail stays on the existing `high` default, while explicit
`original` keeps local images on the original-resolution path.
- Support `high`/`original` consistently for tool image outputs,
including MCP `codex/imageDetail`, code-mode image helpers, and
`view_image`.

codex-rs/exec/src/lib.rs

@@ -637,7 +637,7 @@ async fn run_exec_session(args: ExecRunArgs) -> anyhow::Result<()> {
             let mut items: Vec<UserInput> = imgs
                 .into_iter()
                 .chain(args.images.iter().cloned())
-                .map(|path| UserInput::LocalImage { path })
+                .map(|path| UserInput::LocalImage { path, detail: None })
                 .collect();
             items.push(UserInput::Text {
                 text: prompt_text.clone(),
@@ -657,7 +657,7 @@ async fn run_exec_session(args: ExecRunArgs) -> anyhow::Result<()> {
             let prompt_text = resolve_root_prompt(root_prompt);
             let mut items: Vec<UserInput> = imgs
                 .into_iter()
-                .map(|path| UserInput::LocalImage { path })
+                .map(|path| UserInput::LocalImage { path, detail: None })
                 .collect();
             items.push(UserInput::Text {
                 text: prompt_text.clone(),