feat(agents): render model_policy in agent config + chat banner

[benjackwhite]

What

Mirrors the backend's breaking spec.model → spec.model_policy change (PostHog/posthog#64825) in the desktop agent-applications UI.

• AgentSpec: model: string → model_policy?: ModelPolicy (auto/manual discriminated union), plus ModelEntry / ModelLevel / ReasoningEffort types mirroring the backend schema.
• Model config pane (ModelBody): renders the policy — auto shows policy / level / reasoning; manual shows the primary model + ordered fallback rows (per-model reasoning inline).
• Chat preview banner: new modelPolicySummary() helper → compact label (auto · high, or <model> +N for a manual list).

Display-only — the builder agent still authors specs via its tools; there's no model picker in the desktop UI.

Base

Stacked on posthog-code/m3a-final (#2700) — the agent-applications files only exist there, not yet in main. The merged agent-builder-polish (#2770) also targeted m3a-final.

Testing

• @posthog/shared + @posthog/ui typecheck clean for these files (the only @posthog/web typecheck failures are pre-existing on this branch, in canvas/WebsiteLayout.tsx and code-review/InteractiveFileDiff.tsx — untouched here); Biome clean.
• Verified live against local posthog: patched the local agent-builder live revision to model_policy: { mode: auto, level: high } and confirmed the config pane renders POLICY auto / LEVEL high / REASONING high correctly.

🤖 Generated with Claude Code

[github-actions]

React Doctor found no issues in the changed files. 🎉

_{Reviewed by React Doctor for commit 1b1deaf.}

[greptile-apps]

Prompt To Fix All With AI

Fix the following 1 code review issue. Work through them one at a time, proposing concise fixes.

---

### Issue 1 of 1
packages/ui/src/features/agent-applications/components/AgentConfigurationPane.tsx:640-643
Duplicate-key risk in manual model list. The composite key `${m.model}:${m.reasoning ?? "_"}` is not guaranteed unique — if the same model appears twice in the priority list (e.g., two entries for `"claude-3-5-sonnet"` with no per-entry reasoning), React will emit a duplicate-key warning and reconciliation may misbehave. Array index is a safe alternative for a stable, ordered priority list where items are never reordered in-place.

```suggestion
          {policy.models.map((m, i) => (
            <Row
              key={i}
              label={i === 0 ? "model" : "fallback"}
```

_{Reviews (1): Last reviewed commit: "feat(agents): render model_policy in age..." | Re-trigger Greptile}

[greptile-apps]

Duplicate-key risk in manual model list. The composite key ${m.model}:${m.reasoning ?? "_"} is not guaranteed unique — if the same model appears twice in the priority list (e.g., two entries for "claude-3-5-sonnet" with no per-entry reasoning), React will emit a duplicate-key warning and reconciliation may misbehave. Array index is a safe alternative for a stable, ordered priority list where items are never reordered in-place.

Suggested change

	{policy.models.map((m, i) => (
	<Row
	key={`${m.model}:${m.reasoning ?? "_"}`}
	label={i === 0 ? "model" : "fallback"}
	{policy.models.map((m, i) => (
	<Row
	key={i}
	label={i === 0 ? "model" : "fallback"}

Prompt To Fix With AI

This is a comment left during a code review.
Path: packages/ui/src/features/agent-applications/components/AgentConfigurationPane.tsx
Line: 640-643

Comment:
Duplicate-key risk in manual model list. The composite key `${m.model}:${m.reasoning ?? "_"}` is not guaranteed unique — if the same model appears twice in the priority list (e.g., two entries for `"claude-3-5-sonnet"` with no per-entry reasoning), React will emit a duplicate-key warning and reconciliation may misbehave. Array index is a safe alternative for a stable, ordered priority list where items are never reordered in-place.

```suggestion
          {policy.models.map((m, i) => (
            <Row
              key={i}
              label={i === 0 ? "model" : "fallback"}
```

How can I resolve this? If you propose a fix, please make it concise.