Add tool for filtering currently-active vectors at a given state in a runbook

[RichardRWen]

Added script main/neurips23/runbooks/filter_at_timestep.py

The script takes a runbook, a specific timestep in the runbook, and a binary file of vectors. It determines the active vectors after the given step of the runbook, then outputs a new binary file containing only those active vectors.

Usage: filter_at_timestep.py [-h] -r RUNBOOK -t TIMESTEP -i INFILE -o OUTFILE [-d {float,int,uint8}]
Options:
-h, --help show this help message and exit
-r RUNBOOK, --runbook RUNBOOK
Path to the runbook (.yaml).
-t TIMESTEP, --timestep TIMESTEP
Time step to filter vectors by.
-i INFILE, --infile INFILE
Path to the input binary vector file.
-o OUTFILE, --outfile OUTFILE
Path to the output filtered binary file.
-d {float,int,uint8}, --datatype {float,int,uint8}
Data type of vectors: 'float' (4 bytes), 'int' (4 bytes), or 'uint8' (1 byte).