Regression in plotting of 1d datasets using plot dataset

Consider the example here https://qcodes.github.io/Qcodes/examples/Parameters/Simple-Example-of-ParameterWithSetpoints.html#Measurement before qcodes 0.25 this used to be plotted as a line plot but in 0.26.0 onwards this is now plotted as a scatter plot.

Some debugging indicate that this is because the setpoints and probably also data has gained an extra singleton dimension this means that np.all_close in datatype_from_setpoints_1d will always return True since [[1,2,3,...]] and [1,2,3,...] are always identical due to broadcasting.

Start with the Simple Example of ParameterWithSetpoints example linked in the issue and trace how its 1D setpoints reach datatype_from_setpoints_1d. Check the singleton dimensions involved in the np.allclose comparison; the work is done when the example's 1D dataset is plotted as a line rather than a scatter plot.

Written by the indexing model from the issue text.