refactor(bigframes): Simplify @udf wrapper object

[TrevorBergeron]

Thank you for opening a Pull Request! Before submitting your PR, there are a few things you can do to make sure it goes smoothly:

• Make sure to open an issue as a bug/issue before writing your code! That way we can discuss the change, evaluate designs, and agree on the general idea
• Ensure the tests and linter pass
• Code coverage does not decrease (if any source code was changed)
• Appropriate docs were updated (if necessary)

Fixes #<issue_number_goes_here> 🦕

[gemini-code-assist]

Code Review

This pull request refactors the User Defined Function (UDF) infrastructure by introducing a Udf protocol and a UdfRoutine dataclass, effectively consolidating the handling of BigQuery routines and removing the BigqueryCallableRowRoutine class. Code throughout the library now utilizes isinstance(..., Udf) for type checking. Furthermore, the system tests for managed functions have been significantly updated to use a pytest fixture for more reliable resource cleanup. Feedback was provided regarding the incorrect use of @functools.partial as a decorator on the UdfRoutine.__call__ method, which should be removed.

[gemini-code-assist]

The @functools.partial decorator on the __call__ method of UdfRoutine appears to be incorrectly applied. functools.partial is typically used to create a new callable with some arguments pre-filled, not as a decorator for a method definition. In this context, it is likely either a no-op or could lead to unexpected behavior. Since the __call__ method simply delegates to self.func, this decorator seems redundant and could be removed for clarity.

Suggested change

	@functools.partial
	def __call__(self, *args, **kwargs):