feat: merge media reads into Read tool

[7Sageer]

Related Issue

No linked issue. The problem is described below.

Problem

Read and ReadMediaFile split file reading by content type, so the model has to choose a separate tool for images and videos before the filesystem has actually inspected the file. This makes media reads harder to discover and keeps duplicate read-tool behavior in the runtime, TUI, profiles, permissions, and docs.

Draft note: this PR intentionally starts as draft while the compatibility strategy for the legacy ReadMediaFile tool name is reviewed.

What changed

• Merged image and video reading into Read, including media type sniffing, capability checks, size limits, image dimension summaries, and video upload support.
• Removed ReadMediaFile from the default active tools and default auto-approval policy, while keeping TUI rendering support for old recorded ReadMediaFile sessions.
• Updated TUI chips/renderers so Read media results show compact metadata instead of base64 output.
• Updated English and Chinese tool reference docs, plus tests for text reads, media reads, permissions, turn flow, and TUI media rendering.
• Added a changeset for the CLI package.

Checklist

• I have read the CONTRIBUTING document.
• I have linked a related issue, or explained the problem above.
• I have added tests that prove my feature works.
• Ran gen-changesets skill, or this PR needs no changeset.
• Ran gen-docs skill, or this PR needs no doc update.

[changeset-bot]

🦋 Changeset detected

Latest commit: 545b5d0

The changes in this PR will be included in the next version bump.

This PR includes changesets to release 1 package

Name	Type
@moonshot-ai/kimi-code	Minor

Not sure what this means? Click here to learn what changesets are.

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