feat: add per-conversation effort control

[FrundlesTian]

Description

Adds per-conversation Claude reasoning-effort control now that EffortLevel is available in the Agent SDK.

• adds /effort low|medium|high|xhigh|max, /effort to inspect the current value, and /effort default to remove the override
• forwards the selected effort through the facade to ClaudeAgentOptions for every user-triggered Claude path in agentic and classic modes
• preserves independent effort state for project-topic conversations and exposes it in /status
• updates command menus, help text, README, docs, and changelog

TaskBudget is intentionally out of scope for this PR. Its token ceiling and interaction with the existing max_budget_usd guard should be designed separately before exposing it to users.

Related issue

Related to #233 (implements the agreed effort-only slice; leaves TaskBudget for follow-up).

Type of change

• Bug fix
• New feature
• Breaking change (documented in CHANGELOG under "Changed" or "Removed")
• Documentation or tooling only

How it was tested

• Tests added or updated
• make test and make lint pass locally
• Tested by hand against a running bot

Equivalent local checks run in an isolated Python 3.13 environment:

• pytest --no-cov — 598 passed
• black --check src tests
• isort --check-only src tests
• flake8 src tests
• mypy src/bot/utils/effort.py

The repository-wide mypy src currently reports 540 pre-existing errors on main; this change introduces none in the new helper module.

Checklist

• One concern per PR; unrelated changes are split out
• CHANGELOG.md has an entry under [Unreleased]
• If pyproject.toml dependencies changed, poetry lock was run and the updated poetry.lock is committed (not applicable; no dependency changes)
• Documentation updated (README.md, docs/, .env.example, CLAUDE.md) where settings or commands changed
• New settings default to current behaviour (not applicable; no new setting and unset effort preserves the SDK default)
• If AI tools helped write this change, I reviewed every line and the hand-testing above is mine

[FrundlesTian]

Rebased this branch onto the latest `main` (through #239) to resolve the merge conflict and get a clean re-run of CI/review.

• Conflict was limited to `CHANGELOG.md`'s `[Unreleased]` section (main's CI-fix entries vs. this PR's `Added` entry) — resolved by keeping both.
• No code changes; this is a pure rebase (`9c87a74` → `c642250`).
• Verified locally before pushing: `pytest --no-cov` (598 passed), `black --check`, `isort --check-only`, `flake8`, `mypy src/bot/utils/effort.py`, and `poetry check --lock` — all clean.
• Force-pushed with `--force-with-lease` to update this PR branch.

The earlier `review` check failure was the fork-checkout issue that #239 fixed (this PR predated that fix); it should run cleanly now that the branch is rebased past it.

[github-actions]

No merge-blocking issues found in the current diff.

[FrundlesTian]

Fixed in fd72027: updated CLAUDE.md's classic-mode command count (13 → 15) and added /effort and /restart to the agentic mode commands list at line 133, matching what README.md/docs already had.