docs: document macros as first-class custom commands

Covers top-level .name invocation, the new description/isolated macro
fields (with the non-isolation persistence, fail-fast, nested-macro, and
.exit caveats), workspace .coyote/macros/ + --no-workspace-macros,
enabled_macros scoping at global/role/agent/session levels, and
.macro enable|disable across the README and every example config.
graph.example.yaml gains a note that enabled_macros is ignored in graph
configs.

CHANGELOG intentionally untouched: it is generated by commitizen at
release time from the conventional commit subjects.

Implements plans/custom-commands-design.md §8.

config.agent.example.yaml

@@ -60,6 +60,8 @@ enabled_skills:                  # Optional list of skills available when this a
 inject_skill_instructions: true  # Inject a short hint pointing the model at `skill__list` when skills are enabled
                                  # (default: true). Suppressed automatically when no skills are available.
 skill_instructions: null         # Custom text for the skill hint (optional; uses built-in default if null)
+enabled_macros:                  # Optional list of macros invocable when this agent is active in the REPL.
+  - generate-commit-message      # An empty list disables all macros. Omit to inherit the role/global default.
 memory: null                     # Per-agent memory override (default: inherit). Set to `false` to disable memory
                                  # for this agent regardless of workspace/global presence. See the Memory wiki page.