Improve docs for elwise/op test utilities

[honno]

test_operators_and_elementwise_functions.py is the most complicated part of the test suite. That inherent complexity should be mitigated as much as possible via good docs. I wrote some broader docs in the file, but think docstrings similiar to the one seen in ph.assert_dtype() would be well-suited for the *_assert_against_refimpl() utilities.

array-api-tests/array_api_tests/pytest_helpers.py

Lines 82 to 123 in 87834dd

	def assert_dtype(
	func_name: str,
	in_dtype: Union[DataType, Sequence[DataType]],
	out_dtype: DataType,
	expected: Optional[DataType] = None,
	*,
	repr_name: str = "out.dtype",
	):
	"""
	Assert the output dtype is as expected.
	If expected=None, we infer the expected dtype as in_dtype, to test
	out_dtype, e.g.
	>>> x = xp.arange(5, dtype=xp.uint8)
	>>> out = xp.abs(x)
	>>> assert_dtype('abs', x.dtype, out.dtype)
	is equivalent to
	>>> assert out.dtype == xp.uint8
	Or for multiple input dtypes, the expected dtype is inferred from their
	resulting type promotion, e.g.
	>>> x1 = xp.arange(5, dtype=xp.uint8)
	>>> x2 = xp.arange(5, dtype=xp.uint16)
	>>> out = xp.add(x1, x2)
	>>> assert_dtype('add', [x1.dtype, x2.dtype], out.dtype)
	is equivalent to
	>>> assert out.dtype == xp.uint16
	We can also specify the expected dtype ourselves, e.g.
	>>> x = xp.arange(5, dtype=xp.int8)
	>>> out = xp.sum(x)
	>>> default_int = xp.asarray(0).dtype
	>>> assert_dtype('sum', x, out.dtype, default_int)
	"""