Bug with hwp direction when starting with planet scan

[mmccrackan]

Adding this just to keep track of this until I fix it, but when a schedule starts with a planet scan and when hwp_dir = None in the state, the HWP will always spin in reverse due to the existing logic. For the wiregrid time constant measurements with hwp_dir = 0 in the master schedule, this will be a problem since they rely on the current spin direction.