handle images from burst mode scans

[DanPorter]

Example scan on I16 using burst mode:
'/dls/i16/data/2025/mm39624-2/1102152.nxs'

With attributes:

cmd = scan phi 0 10 1 pil3_100k_burst 0.1
axes = /entry/measurement/phi
signal = /entry/measurement/count_time
shape = (11,)
image_path = /entry/instrument/pil3_100k_burst/data
image_shape = (11, 10, 195, 487)

So the image shape (2 scannable axes) is different from the axes and signal (1 scannable axes).

Using the current methodology works, but returns a stack of images:

import hdfmap
m = hdfmap.create_nexus_map('/dls/i16/data/2025/mm39624-2/1102152.nxs')

scan_length = m.scannablers_length()  # = 11
index = m.get_image_index(scan_length // 2)  # = (5,)
image = m.get_image(m.load_hdf(), index)  # array[10, 195, 487]

it is unclear what should happen in this case - should the image stack be summed, averaged or just the first entry be given?

Should m.get_image always return a 2D array?