Fix tuple type support in typecheck() and flatmap inference (fixes #632)

[Zeyad-Hassan-1]

Fixes #632

What was broken:

1. Tuple Type Validation: typecheck() in types.py was not handling tuple types correctly. When a user passes (str, int) as a type, get_origin() returns None, so the existing isinstance(input_type, Tuple) check never matched.

   • Fix: Added a separate branch using isinstance(input_type, tuple) with recursion to support nested types like (typing.Tuple[str, int], typing.Tuple[str, int]).

2. Python 3.12+ Compatibility: get_type_flatmap_function() used type(sig.return_annotation) != type(Iterable), which breaks in Python 3.12+ because Iterable and Iterable[str] are represented as different internal classes.

   • Fix: Switched to using get_origin(sig.return_annotation) is not collections.abc.Iterable for more robust validation across Python versions.

Verification:

The previously skipped test_wordcount now passes end-to-end, integrated with the Java JSON server.

Note: During testing, I observed that the word count results are consistently off by +1 for each word.

[Zeyad-Hassan-1]

Hi @zkaoudi , could you please approve the CI workflows so the checks can run? I'd also appreciate any feedback on the implementation. Thank you!

[mspruc]

For some reason the actions workflows are not triggering for this PR, ill try to convert to draft to see if it helps

[mspruc]

weird, closing and reopening helped, your workflows are running

[mspruc]

Ty for your contribution