docs(brainstorm-with-docs): уточнены границы артефактов по внешнему ревью

- Зачем:
  - внешнее ревью выявило риск двойной записи решений (spec/ADR) и
    терминологическую путаницу; правки документные, поведение не меняют.
- Что:
  - adr.md и spec.md: добавлен lifetime-критерий ADR vs spec (ADR — для
    решений, переживающих спеку; локальные остаются в спеке).
  - spec.md: разведены Options considered (до выбора) и Decisions (после).
  - SKILL.md: блок <validation> переименован в <self_review>, чтобы не
    путать с секцией ## Validation в spec.md.
  - README.md: честнее описано авто-срабатывание на Codex.
  - SKILL.md: у SEAM зафиксирован TODO(F1) про строку дивергенции
    (отложена ради чистого baseline на обкатке).
- Проверка:
  - python3 yaml-парс frontmatter; нет висячих ссылок на <validation>.

agents/.agents/skills/brainstorm-with-docs/artifacts/spec.md

@@ -40,6 +40,13 @@ add nothing for this particular problem — an empty section is noise.
 
 ## Notes on key sections
 
+**Options considered vs Decisions.** Keep these two sections doing different jobs
+so the rejected options aren't written twice. *Options considered* is the
+landscape *before* the choice — the candidates and their trade-offs as they
+stood. *Decisions and rejected alternatives* is the outcome *after* the choice —
+what was picked and why each alternative was rejected. Frame in the first, judge
+in the second.
+
 **Decisions and rejected alternatives** is the highest-value section. Capture:
 
 - accepted decisions and the reason behind each
@@ -48,9 +55,10 @@ add nothing for this particular problem — an empty section is noise.
 - sequencing decisions
 - assumptions future agents must not silently re-derive
 
-Do not turn this section into an ADR dump. Promote a decision to its own ADR
-(`artifacts/adr.md`) only when it is durable, surprising, and the result of a
-real trade-off.
+Do not turn this section into an ADR dump. Keep decisions that are local to this
+work here; promote a decision to its own ADR (`artifacts/adr.md`) only when it is
+durable, surprising, the result of a real trade-off, and will still matter after
+this spec is obsolete (see the lifetime test in `artifacts/adr.md`).
 
 **Validation** describes how we will know the implementation is correct — tests,
 checks, acceptance criteria. It belongs here, not in `CONTEXT.md`.