New interchange `target_density` scale factor behaviour does not have a direct 1.1 relationship with previous behaviour

[IAlibay]

Description

The current news item for the scale factor change states that "previous behavior can be restored by passing scaling the target_density argument down by a factor of 1.1" (note there's a typo in the news item!).

However, this is incorrect - it's actually closer to 1.33. I've not been able to nail down the exact number because of the ** (1/3), but 1.33 is close enough. (Edit: actually, I think it is just 1.1**3 - someone double check my maths though)

Reproduction

Assume a total_mass of 100, and an initial_volume of 50.

Scale of 1.1 and target density fo 0.95

volume (as passed to _scale_box) = 100 / 0.95 = 105.2632
volume_scale_factor = 2.1053
linear_scale_factor = 1.4089

Scale of 1.0 and target density of 0.8636 (~1.1 scaled down)

volume (as passed to _scale_box) = 100 / 0.8636 = 115.7943
volume_scale_factor = 2.3159
linear_scale_factor = 1.3230

Scale of 1.0 and target density of 0.715 (~1.33 scaled down)

volume (as passed to _scale_box) = 100 / 0.715 = 139.8601
volume_scale_factor = 2.7972
linear_scale_factor = 1.4089

Software versions

Current develop version of interchange.