Improve aimer trapezoidal profile logic

[jkleiber]

Purpose
The purpose of this addition is to address any aimer control deficiencies uncovered that are not deemed to be bad enough to fix in integration prior to VABLA.

List of such potential problems:

• Currently we are winding up the trapezoidal profile timer by never resetting it in the PRIME state. After a period of time we should expect the control to always issue step commands. Maybe this is fine, but it does not seem like the cleanest implementation.

Project Scope

• Clean up the implementation of the trapezoidal profile for the aimer

[jkleiber]

Additional Info for the future: The behavior seen in #111 (comment) was likely caused by turning the PID controller off when near the setpoint rather than anything related to the profile improvements proposed in the scoring PR themselves.

We should still wait to get the integration branch merged before doing this change, but I wanted to leave this note that we aren't broken if we do this change in the future.