feat(agent-designer): reflect governed agent bindings in the chat-input (lock pickers)

[philmerrell]

Summary

Closes the frontend half of the Agent binding-resolution work. The backend now governs an Agent's model/tool/skill bindings at invocation (#601, #602) — the agent's set wins regardless of what the client sends. But the chat-input still showed the model/tool/skill pickers as free-select, which was dishonest: the user could "change" a model the backend would ignore. This locks each picker to the active Agent's bindings, per primitive.

This is UI honesty, not enforcement — the backend remains the authority. Per-primitive: an agent that binds a model but no tools locks only the model; the rest stay free-select. Plain chat and legacy assistants are unchanged.

What changed

• Session page (session.page.ts): inject AgentService/ToolService/SkillService; fetch the governed Agent alongside the assistant (agentId == assistantId) in loadAssistant; apply per-primitive locks from modelConfig/bindings, and release them when navigating to plain chat. Best-effort: the /agents surface may be disabled (404) or the assistant may be a legacy assistant with no bindings — every failure leaves the pickers free-select and never blocks the conversation.
• ModelService / ToolService / SkillService: small agent-lock API (lockToAgent* / clearAgentLock + agentLocked/agentModelLocked). While locked, enabledToolIds/enabledSkillIds return the bound set (replace semantics, matching the backend), toggles no-op, and isToolShownEnabled/isSkillShownEnabled render the bound set honestly (greyed).
• UI: model-dropdown shows a locked read-only chip; model-settings shows a "Set by agent" model row and a "This agent uses a fixed set of tools/skills" banner, with tool/skill/sub-tool toggles disabled while locked.

Tests

• +5 tool-lock, +5 skill-lock, +4 model-lock service specs (ng test, 51 pass in the affected specs)
• tsc clean; production build (AOT template validation) clean
• No local preview (needs backend + AGENTS_API_ENABLED + auth) — see the manual test script in the PR discussion.

Known limitations (follow-up)

• Model race: if the pinned model isn't in the user's loaded set yet, the dropdown disables but may show the fallback name until models load. (Models load on app init, so rare.)
• Skill-lock banner only renders in skills chat-mode (the skills section is mode-gated); the backend still forces skill-mode + the bound skills regardless.

🤖 Generated with Claude Code