Modernize generator tests

[DhanushPillay]

Proposed Changes:

This PR modernizes the test/components/generators/ test suite by fully adopting strict mypy type hints.

I've gone through the generator tests and cleaned them up to make sure they're completely type-safe. Specifically, this includes:

• Adding explicit return types (like -> None) and parameter types (e.g., using Any for fixtures and monkeypatches) across all the root generator and chat generator tests.
• Refactoring assertions that previously relied on str | None so they are now strict-type compliant (e.g. safely checking .text).
• Doing a full sweep to remove any redundant or unused # type: ignore comments that were cluttering up the codebase.
• Finally, I've added the test/components/generators/ directory to the types check in pyproject.toml so these tests will be officially checked in the CI pipeline going forward.

How did you test it?

I verified everything locally by running hatch run test:types. The entire test/components/generators/ directory now passes with zero errors.

Notes for the reviewer

Since this is purely an internal update to the test suite and build config (no user-facing API or feature changes), I skipped generating a release note as per the contribution guidelines for CI/test changes.

[vercel]

@DhanushPillay is attempting to deploy a commit to the deepset Team on Vercel.

A member of the Team first needs to authorize it.

[CLAassistant]

All committers have signed the CLA.

[DhanushPillay]

Hey @sjrl , thanks for the review! I just pushed a commit with those updates.

I've added the specific error codes to all the type: ignore comments (like [attr-defined]) so they're no longer generic. For the message.text checks, I dropped the ignores and just added assert message.text is not None before using them, as you suggested. I also fixed a quick merge conflict in pyproject.toml and ran the formatter so the CI should be green now.

Let me know if everything looks good!

[github-actions]

Coverage report

This PR does not seem to contain any modification to coverable code.

[HaystackBot]

Hi @DhanushPillay, thanks a lot for your contribution! 🙏

We noticed that the Contributor License Agreement (CLA) check (license/cla) hasn't passed yet, so we've temporarily moved this PR to draft and paused the review assignment.

To get your PR reviewed, please sign the CLA via the link in the license/cla check below (or in the CLA bot comment). As soon as the check turns green, this PR will automatically be marked ready for review again and a reviewer will be re-assigned.

[sjrl]

@DhanushPillay sorry false positive with the bot.

[DhanushPillay]

@sjrl I have just pushed a commit to address the remaining review comments. I added the missing return types, replaced the type ignores in test_fallback.py with strict type hints, and kept the dynamic __haystack_*__ attributes as expected. I also ran ruff format to ensure all formatting is correct. Please let me know if anything else is needed

[sjrl]

Do you know why this one needs an ignore? The type for tool_result is ToolCallResultContentT = str | Sequence[TextContent | ImageContent | FileContent] which I thought matches the input correctly.

[DhanushPillay]

You're absolutely right, it doesn't need the ignore! The issue was that without an explicit type annotation on the tool_result list, mypy inferred a narrower type (list[TextContent] or similar) instead of recognizing it as a valid Sequence[TextContent | ImageContent | FileContent]. I've fixed it by adding an explicit type hint on the variable, so mypy can see it matches ToolCallResultContentT directly. Thanks for catching that

[DhanushPillay]

@sjrl Hey, I've addressed all the review feedback and CI looks good on my end, the only failing check is the Vercel deploy which seems unrelated. Happy to make any other changes if needed

[sjrl]

Thanks!