Align pool to batch during scale up

[rbtr]

Signed-off-by: Evan Baker rbtr@users.noreply.github.com

Reason for Change:

Currently the scale-up logic blindly adds +1 batch to whatever the current request is, meaning that if the current request is not aligned to the batch, the post scale-up pool will still not be aligned to the batch. During scale down, we calculate the skew from the batch and use that to realign the pool to a multiple of the batch.
This change copies the mod arithmetic from the scale-down so that we realign to the next greater multiple of the batch size during a scale-up. This may be within the threshold that triggers another scale-up, ie:
a pool of batch 16 with 15 IPs and 14 assignments will scale up to -> 16, which is less than the minimum free IPs, so it will scale up again to 32 on the next tick. This behavior is seen when coming out of a constrained subnet state, when the batch would have been temporarily adjusted to 1 and thus the requested IP count could be anything.

Issue Fixed:

Requirements:

• uses conventional commit messages
• includes documentation
• adds unit tests

Notes:

[rsagasthya]

Approving the changes. But, looks like the branch is out of date.