fix(ollama): verify model after pull

[Sanjays2402]

Summary

Ollama pulls now report success only after the model appears in the installed-model list. The adapter retries the visibility check once after a short backoff and returns a failed PullResult if the model remains absent.

Closes #15

Changes

• Verify post-pull model visibility before completing progress or reporting success.
• Cover delayed visibility and permanent absence with deterministic unit tests.

Testing

• pytest -q -m 'not integration' (91 passed)
• Ruff check/format, Bandit, Mypy, and the integration skip path without Ollama

Checklist

• Branch named per Git Workflow (feature/, fix/, docs/, refactor/, test/, chore/)
• Not developed on main
• Code follows AGENT.md coding standards (type hints, Pydantic v2, no generic Exception, no business logic in CLI)
• domain/ and ports/ stay pure (no I/O, no framework imports)
• Quality gates pass locally: ruff, mypy --strict, bandit, pytest
• Docs updated if behavior changed (not needed; no public API or configuration changed)
• pyproject.toml and src/modeldock/__init__.py versions match (not release-related)

[github-actions]

🎉 Congratulations @Sanjays2402!

Your pull request has been successfully merged into main. 🚀

Thank you for contributing to OpenAgentHQ and helping improve the project.

We truly appreciate your contribution and hope to see you back with more amazing PRs!

Happy Open Sourcing! ❤️

🌟 This is your first merged contribution to this repository.
Welcome to the OpenAgentHQ contributors family!