Clarify CodingScaffold value proposition

[JRS1986]

Agentic coding disclosure

• Agent / tool used: Codex
• Model / provider: OpenAI GPT-5
• Human operator: @JRS1986
• Task: Make the README explain why teams should use CodingScaffold when coding agents already install with one command. Closes Clarify why teams should use CodingScaffold #82. Follow-up design captured in Design chat-to-knowledge ingestion around abstraction, not raw transcripts #83.

What changed

• Files changed: README, docs homepage, FAQ, Knowledge Base docs, changelog, docs-positioning regression test.
• Commands run: uv run ruff check; uv run pytest -q; uv run coding-scaffold context lint --target .; uv run coding-scaffold eval run --target .
• Tests run: uv run ruff check passed; uv run pytest -q passed with 333 tests; context lint passed; eval passed 9/9.
• Tests not run and why: n/a.

Risk surface

• External tools or MCP servers used: GitHub CLI for issue, branch push, and PR operations.
• Secrets / data exposure risk: none.
• Network calls made during the session: GitHub issue, branch push, and PR operations only.

Review focus

• Human review focus: confirm the README answers "why not just install an agent?" early enough; confirm knowledge wording does not overclaim automatic chat ingestion; confirm future ingestion guidance emphasizes distilled reviewed proposals instead of raw transcript storage.
• Known limitations: no automatic chat-ingestion feature was implemented; Design chat-to-knowledge ingestion around abstraction, not raw transcripts #83 tracks future design constraints.
• Reusable knowledge captured: Clarify why teams should use CodingScaffold #82 records the README value-proposition feedback; Design chat-to-knowledge ingestion around abstraction, not raw transcripts #83 records the chat-ingestion abstraction constraint.