docs: refresh bilingual readmes and agent benchmark figures

[LivXue]

Summary

Present Raven's orchestration capabilities and four specialized agents with benchmark figures, and keep the English and Chinese READMEs aligned.

• Add the Multi-Agent Orchestration Benchmark at the end of the introduction. Compare Hermes, Claude Code, and Raven with shared 0-1 scales for Node F1, Edge F1, Partial Order Accuracy, and Exact Match Rate.
• Move Raven Agents before Benchmarks and add concise Research, Code, Design, and Oncall descriptions with their figures.
• Replace the third-party agent table with a centered card image and explain how Raven connects to and orchestrates these agents. Host image files as GitHub attachments.
• Standardize primary heading markers and align section order, captions, model settings paths, and the EverMind ecosystem catalog across both languages.

Type

• Fix
• Feature
• Docs
• CI / tooling
• Refactor
• Other

Verification

Commands run after rebasing onto the latest main:

uv run --no-sync pytest tests/test_readme_scope_canon.py tests/test_living_docs.py -q -o addopts=''
# 4 passed
uv run --no-sync pre-commit run --files README.md README.zh-CN.md
# All applicable checks passed
make check-large-files
# Passed
make check-source-language
# Passed
env PYTHONPATH=. uv run --no-sync python scripts/check_commit_messages.py origin/main..HEAD
# Passed
git diff --check
# Passed

Compared all 27 headings, 12 code blocks, image URLs and sizing, commands, paths, numeric values, and architecture graph structure across both languages. Verified all 24 orchestration scores against the supplied results and confirmed that the uploaded chart matches the local PNG byte for byte.

• Relevant tests pass locally
• Relevant lint / type checks pass locally
• User-facing docs or screenshots are updated when needed

Risk

This changes documentation structure, wording, and linked figures. Runtime behavior is unchanged. Figures depend on GitHub attachment availability. Roll back by reverting the changes to README.md and README.zh-CN.md.

• Security impact considered
• Backward compatibility considered
• Rollback path is clear for risky changes

Related Issues

#462, #465, #474, #477, #479

[gloryfromca]

No blockers; this can merge as far as I am concerned.

Reviewed the full github/main...HEAD diff at c2b7ed0f48cd, including the follow-up that expands the two single-column charts. I covered the repository rules in AGENTS.md, canonical architecture terminology from CONTEXT-MAP.md/CONTEXT.md, bilingual structure and content parity, GitHub rendering, all newly added external and attachment URLs, surrounding README/UI/agent references, commit history, backward compatibility, and test integrity. This is documentation-only, changes no runtime callers or architecture, and does not weaken or remove tests.

Validation: uv run pytest tests/test_readme_scope_canon.py tests/test_living_docs.py tests/test_docker_runtime.py -q -o addopts='' passed (13 tests); the large-file and source-language scripts passed; git diff --check github/main...HEAD passed. GitHub's renderer preserved the matched image/heading structure in both READMEs and the new full-width attributes.

[gloryfromca]

Blocking: restore the Quick Start contract and replace the unreadable AI4S chart.

I reviewed the new chart-layout commit and the full rebased github/main...HEAD diff. I covered the current AGENTS.md and CONTEXT terminology rules, the rebase and commit history, GitHub-facing document structure and assets, the installer test that consumes the README, backward compatibility, and test integrity. Runtime architecture and callers are unchanged, and this PR does not modify or weaken tests.

Validation: test_readme_quickstart_matches_the_installer fails locally with StopIteration, matching the current CI shard failure; the other focused README/living-doc/Docker tests pass (13 tests). The large-file and source-language checks and git diff --check pass.

[gloryfromca]

No blockers; this can merge as far as I am concerned.

Reviewed the a0cedc452 test-only delta and the full current PR diff. The semantic, localized Quick Start lookup fixes the prior CI failure while preserving the three installer-content assertions and adding exact-one-section validation. The author's native-resolution evidence, independently reproduced from identical asset bytes, refuted my AI4S finding; I withdrew it and resolved both threads I opened.

Covered the current test caller and contract, AGENTS.md language rule, commit history, backward compatibility, test integrity, and the unchanged runtime/architecture surface. Validation: the full onboarding test file passed (332 tests); focused README/living-doc/Docker tests passed (13 tests); large-file, source-language, and git diff --check checks passed.

[gloryfromca]

No blockers; this can merge as far as I am concerned.

Reviewed the new chart-rotation delta and the full current PR comparison. Both replacement URLs resolve to valid 2000 px-wide PNGs; native-resolution comparison confirms every prior method, model, category, and value is preserved, while labels, legends, axes, and values remain unclipped and non-overlapping. The English and Chinese READMEs still use identical assets and links.

Covered AGENTS.md/CLAUDE.md rules, the diff, README rendering consumers and installer-test caller, commit history, backward compatibility, test integrity, and the unchanged runtime/architecture surface. Validation at this head: 345 README/onboarding/living-doc/Docker tests passed; source-language, large-file, and whitespace checks passed; all completed GitHub checks pass (one independent unit shard was still running when reviewed). Patch-risk result: merge, low impact and regression likelihood, strong changed-path protection, easy revert, high confidence.