UL: Top-Tagging

[mcremone]

For top-tagging, ParticleNet is used already in the current mono-top analysis. Two items need to be implemented:

1. Top-tagging working points
2. Top-tagging efficiency scale factors and mis-tagging efficiency scale factors.

For item 2) a function needs to be implemented in corrections.py, following the example of the double-b tagging scale factors:

https://github.com/mcremone/decaf/blob/UL/analysis/utils/corrections.py#L521