fix(skills): use utility model for skill tests instead of chat default

[holden093]

Summary

Skill tests were using resolve_endpoint("default", ...) which returns the chat model. They should use the utility model, like auto-naming and memory audit already do (fixed in PR #4027). One-word change: "default" → "utility".

Note: Re-opening of #5026, which was auto-closed on 2026-07-23 when the repository was transferred to the odysseus-dev org (the dev history rewrite orphaned the old base and forced all open PRs closed). This branch is the same one-line fix cleanly rebased onto the current dev. Original review thread: #5026.

Target branch

• This PR targets dev, not main.

Linked Issue

Fixes #5025

Type of Change

• Bug fix (non-breaking — fixes a confirmed issue)

Checklist

• I searched open issues and open PRs — this is not a duplicate.
• This PR targets dev
• My changes are limited to the scope described above — no unrelated refactors or whitespace changes mixed in.
• I verified the change by running a skill test and confirming the logs show the utility model instead of the chat model.

How to Test

1. Configure different models for Default Chat and Utility in Settings
2. Go to Skills, pick a skill, click Test
3. Check logs: the skill test agent round should show the utility model, not the chat model

Visual / UI changes

None — backend-only change.

[holden093]

Heads-up: the failing Python tests (pytest) check is a pre-existing dev bug, not this PR

The pytest job fails at test collection, before any of this PR's changes are exercised:

NameError: name 'Any' is not defined

Root cause is on dev itself: src/agent_loop.py annotates with dict[str, Any] (lines 1503 and 1877) but only imports AsyncGenerator, List, Dict, Optional, Set — no Any, and no from __future__ import annotations. So importing the module raises NameError, which breaks collection for every test that transitively imports agent_loop (hence the repo-wide failure).

This PR doesn't touch agent_loop.py; every other check (syntax, CodeQL, lint, security scans) passes here, and this PR's own tests pass locally.

One-line fix (add Any to the typing import) is included in #5751. Once dev carries that fix, this PR's pytest will go green with no changes needed here.