Fixed TypeError being thrown on certain Gaussian states in the Plotter.plot_tracks function

[gawebb-dstl]

For some Gaussian states the uncertainity can't be plotted by the Plotter class. You can see if you try running the following code

from stonesoup.plotter import Plotter
from stonesoup.types.track import Track
from stonesoup.types.state import GaussianState, StateVector

track = Track([GaussianState(state_vector=[0, 0], covar=[[10, -1], [1, 10]])])
Plotter().plot_tracks(track, mapping=[0, 1], uncertainty=True)
# TypeError: ufunc 'arctan2' not supported for the input types, and the inputs could not be safely coerced to any supported types according to the casting rule ''safe''

This is caused by complex eignevalues or eigenvectors from the uncertainity matrix. I've put in a check for these in the Plotter class. States which cause complex eignevalues or eigenvectors are now skipped in the plot_tracks function and a warning is raised. A test has been created for this behaviour

[gawebb-dstl]

I think the same error is present in the Plotterly class but haven't tested it