SlopesFF with 352cells for correcting different cell conditions

This is related to https://git.xfel.eu/calibration/planning/-/issues/107

Description

As it turns out SlopesFF doesn't depend on operating memory cells. There is a plan to only inject FF constants for all 352 cells.

This MR:

1. Adds a deviation during injection of these constants.
2. Slicing the retrieved SlopesFF constant based on the number of operating memory cells.

How Has This Been Tested?

This was tested by injecting to max-exfl017 SlopesFF with a low deviation of 352 and then later retrieving them while asking for a constant with 64 memory cells.

https://in.xfel.eu/test_calibration/calibration_constants/20487#condition

Relevant Documents (optional)

Types of changes

• New feature (non-breaking change which adds functionality)

Checklist:

Reviewers

@schmidtp @hammerd