Fix pattern generator tests

[subinsaji]

Fixes #432

[subinsaji]

There is a lot of overlapping logic with hdf_writer.py :

ophyd-async/src/ophyd_async/epics/areadetector/writers/hdf_writer.py

Lines 81 to 112 in 0269697

	self._datasets = [
	_HDFDataset(
	data_key=name,
	dataset="/entry/data/data",
	shape=frame_shape,
	dtype_numpy=dtype_numpy,
	multiplier=multiplier,
	)
	]
	# And all the scalar datasets
	for ds_name, ds_path in self._scalar_datasets_paths.items():
	self._datasets.append(
	_HDFDataset(
	f"{name}-{ds_name}",
	f"/entry/instrument/NDAttributes/{ds_path}",
	(),
	"",
	multiplier,
	)
	)
	describe = {
	ds.data_key: DataKey(
	source=self.hdf.full_file_name.source,
	shape=outer_shape + tuple(ds.shape),
	dtype="array" if ds.shape else "number",
	dtype_numpy=ds.dtype_numpy,
	external="STREAM:",
	)
	for ds in self._datasets
	}
	return describe

and also with panda _hdf_writer.py :

ophyd-async/src/ophyd_async/panda/writers/_hdf_writer.py

Lines 69 to 79 in 0269697

	describe = {
	ds.data_key: DataKey(
	source=self.panda_device.data.hdf_directory.source,
	shape=ds.shape,
	dtype="array" if ds.shape != [1] else "number",
	dtype_numpy="<f8", # PandA data should always be written as Float64
	external="STREAM:",
	)
	for ds in self._datasets
	}
	return describe

From a DRY point of view, we can reduce this repeating logic