Add initial AI-agents configuration

[Oleg-Melnik]

What

Adds the initial AI-agents configuration for the Chords repository, defining
the operating policy, project context, and task-specific skills that agent
tools (Claude Code, OpenAI-based interfaces, etc.) pick up when working in
this codebase.

Contents

• AGENTS.md — the repository operating policy: Git-history and safety
  rules, versioning/report requirements, verification commands for both
  toolchains (root project on JDK 11 / Gradle 6.9.4 and codegen/plugins on
  JDK 17 / Gradle 9.4.x), development conventions, and code-review guidance.
• CLAUDE.md — a thin pointer to AGENTS.md for Claude-based tools.
• .agents/project.md — project overview, module map, architecture notes
  (class-based component model, codegen flow), documentation ownership, and
  CI notes.
• .agents/skills/ — a skill index plus ten task-specific skills, each with
  a SKILL.md policy and agents/openai.yaml UI metadata:
  • engineer — router for mixed or unclear implementation work;
  • component-engineer — class-based Compose UI components across core,
    proto, and client;
  • codegen-engineer — ProtoData plugins, codegen runtime, and Protobuf
    declarations;
  • build-engineer — Gradle build logic, publishing wiring, and version
    policy;
  • tester — test authoring and verification strategy;
  • code-reviewer — implementation review;
  • security-reviewer — credentials, workflow secrets, dependency
    provenance, and agent prompt/configuration safety;
  • ci-engineer — GitHub Actions workflows;
  • docs-writer / docs-reviewer — documentation authoring and review.

Other changes

• Bumped the version to 2.0.0-SNAPSHOT.88 and regenerated pom.xml and
  dependencies.md.