Add capability for computed kadi states and use for attitude and sun-related states

[taldcroft]

Description

The prior implementation of a number of computed quantities was too tightly coupled to the ManeuverTransition and it was generally brittle and not easy to expand.

This PR recognizes that some state values like ra, dec, roll can be computed from other states (q1..4).

The implementation of this is straightforward, with the exception of dealing with the transition every 10ks in NPM to inject a state boundary and force updates to all the sun-angle values.

TODO: There is some intricacy in the code here that still needs better comments and possible cleaning.

This update changes the way that maneuver states are computed. Previously the attitude quaternion corresponded to the start of each state. This is not ideal and there was a comment suggesting that it be changed. Now the quaternion corresponds to the attitude at the midpoint of the state.

Previously there was some hackery to get the pitch, off_nom_roll to match more closely the midpoint value by averaging the start/stop values. Now the values correspond to correct mid-point values for all state bins. This is possible because the values are all computed post-facto when the state start/stop times are known.

Interface impacts

Attitude and sun-related state values will change slightly and for the better.

Testing

Unit tests

• No unit tests
• Mac
• Linux
• Windows

Independent check of unit tests by [REVIEWER NAME]

• [PLATFORM]:

Functional tests

No functional testing.