docs: integrate BDD lifecycle practice

[dapi]

What changed

• add a canonical Behavior Specification Practice for the BDD cycle Discovery → Formulation → Automation;
• integrate behavior discovery and scenario-quality gates into Feature Flow;
• add downstream behavior coverage and change-impact rules to Use Case Flow;
• extend feature brief, use-case, design, implementation-plan, and feature-local use-case templates;
• clarify BDD automation boundaries in testing policy;
• update priming manifests, indexes, artifact routing, and README navigation;
• cite Dan North's original Introducing BDD article as a methodology source.

Why

Memory Bank already separates project-level UC-*, feature acceptance SC-* / NEG-*, checks CHK-*, and test evidence, but did not define how BDD should operate across those owners. Without an explicit contract, downstream projects can duplicate requirements in Gherkin, treat BDD as an E2E framework, or create a parallel scenario hierarchy.

This change keeps one ownership chain:

UC → BR/REQ → SC/NEG → CHK → automated test → EVID

BDD remains a practice inside the selected delivery flow, not a new Task Routing branch or a memory-bank/bdd/ catalog.

Impact

• behavior-heavy features discover rules, examples, questions, and deferred scope before Problem Ready;
• structured examples use domain language and observable context/event/outcome;
• plans map required scenarios to the lowest reliable test surface rather than defaulting to E2E;
• use cases remain canonical for stable project behavior while feature briefs retain acceptance ownership.

Validation

• ruby tools/validate-priming-manifests.rb template/memory-bank
• memory-bank-cli lint --scope-root template/memory-bank --entrypoint template/memory-bank/README.md
• memory-bank-cli doctor --profile template
• git diff --check

Closes #113