feat: custom instructions and environment-aware shell context

[dorlanpabon]

Issue for this PR

Closes #47011

Type of change

• New feature

What does this PR do?

Native port of #47012 to the v2 branch (which restructured config, instructions, settings and docs, so a direct cherry-pick was impossible):

• Free-text customInstructions in config schema (v1 + migrate + normalize), merged global-first and injected as <custom_instructions> after env/date and before AGENTS.md discovery.
• V2 <env> block gains human-readable OS and resolved effective Shell (respects config/shell plugin + discovery); execution untouched.
• Settings editor (own section + controller, en-only i18n) and docs (config.mdx, instructions.mdx ordering).

Note: sync.updateConfig is currently stubbed upstream (TODO: Restore config updates), so the editor wiring typechecks but persistence lands once that TODO is restored.

How did you verify your code works?

• bun run typecheck in schema, core, app, www: pass
• core: instructions builtins (7), config migrate: pass
• app: settings general controllers: pass
• session-generate + session-runner suites: 187 pass

Screenshots / recordings

New settings section only; happy to add a screenshot on request.

Checklist

• I have tested my changes locally
• I have not included unrelated changes in this PR

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

[github-actions]

The following comment was made by an LLM, it may be inaccurate:

Potential Related PR Found:

• PR feat: custom instructions and environment-aware shell context #47012 - feat: custom instructions and environment-aware shell context #47012 (feat: custom instructions and environment-aware shell context)

Why it's related: PR #47012 is the original feature implementation that PR #47043 is based on. According to the PR description, #47043 is a native port of #47012 to the v2 branch (due to restructured config, instructions, and settings that made a direct cherry-pick impossible).

Note: PR #47012 appears to be on a different branch, so these are intentionally related but on separate branches rather than true duplicates.