docs: align cli option behavior

[tompsantos]

Summary

Aligns the CLI option documentation with the current validation and prompt-mode behavior.

The configuration guide currently contains outdated conflict rules and recommends combining --yolo with -p, although prompt mode already uses auto permission and rejects that combination.

Changes

• update the documented option conflict rules
• clarify that permission and Plan modes can override resumed session state
• replace the invalid kimi --yolo -p example
• document KIMI_MODEL_OUTPUT_FORMAT and its precedence
• update the English and Chinese documentation together

Validation

• checked against apps/kimi-code/src/cli/options.ts
• checked against apps/kimi-code/test/cli/options.test.ts
• documentation-only change
• no changeset required

[changeset-bot]

⚠️ No Changeset found

Latest commit: 426a365

Merging this PR will not cause a version bump for any packages. If these changes should not result in a new version, you're good to go. If these changes should result in a version bump, you need to add a changeset.

This PR includes no changesets

When changesets are added to this PR, you'll see the packages that this PR includes changesets for and the associated semver types

Click here to learn what changesets are, and how to add one.

Click here if you're a maintainer who wants to add a changeset to this PR

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 426a3650ff

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Document resumed mode overrides as persistent

For interactive resumes, these flags are not just temporary: applyStartupModesToResumedSession calls session.setPermission('auto'/'yolo') and session.setPlanMode(true) with no exit-time restore, and those session APIs write persisted permission/plan records. A user following this sentence could run kimi --continue --auto, exit, and later resume the same session still in auto mode despite expecting the override to disappear; the zh mirror has the same wording, so update both copies together.

AGENTS.md reference: docs/AGENTS.md:L19-L21

Useful? React with 👍 / 👎.