Allow for time-dependent observer location

[wtbarnes]

Currently, the observer location of an instrument is specified when the instrument is instantiated and is assumed to be the same at each timestep. This assumption should be relaxed to allow for varying observer location such that the SkyCoord describing the observer can either have length 1 or the same length as the number of simulation timesteps (interpolated to the instrument cadence) that is being synthesized. This will involve recalculating the pixel grid at each timestep as the HPC frame will be different each time.