feat(checks): accept model strings in set_default_generator

[henchaves]

Description

set_default_generator now accepts a model identifier string as well as a BaseGenerator instance. A string is wrapped in giskard.agents.Generator, so the common case no longer needs a separate import:

from giskard.checks import set_default_generator

set_default_generator("azure/gpt-5.6-luna")

Passing a generator instance is unchanged, including identity:

from giskard.agents import Generator
from giskard.checks import set_default_generator

set_default_generator(Generator(model="azure/gpt-5.6-luna"))

The string is converted with the existing giskard.agents.Generator import. A lazy from giskard.agents.generators import Generator is not used because giskard-checks may only import that library from the package root.

Also regenerates THIRD_PARTY_NOTICES.md (boto3/botocore 1.43.78 → 1.43.79) so make check-notices matches current licensecheck output.

Related Issue

N/A — requested as an API convenience.

Type of Change

• 📚 Examples / docs / tutorials / dependencies update
• 🔧 Bug fix (non-breaking change which fixes an issue)
• 🥂 Improvement (non-breaking change which improves an existing feature)
• 🚀 New feature (non-breaking change which adds functionality)
• 💥 Breaking change (fix or feature that would cause existing functionality to change)
• 🔐 Security fix

Coding agents

Autonomous agents with no human in the loop must read AUTONOMOUS.md before opening a PR.

PR title: agent-opened PRs must end the title with 🤖🤖🤖🤖 (exactly four robot emojis). Do not omit — that suffix is how the expedited agent PR workflow picks up the PR.

Checklist

• I've read the CODE_OF_CONDUCT.md document.
• I've read the CONTRIBUTING.md guide.
• I've written tests for all new methods and classes that I created.
• I've written the docstring in NumPy format for all the methods and classes that I created or modified.
• I've updated the uv.lock running uv lock (only applicable when pyproject.toml has been modified)