feat(script): Use extracted data for manuscript

For the first time we use the actual final extracted data from relevant studies
to do analysis on instead of just the intermediate Zotero-provided metadata.

We still inject the intermediate metadata where it may be useful (things like
citation counts and keywords) but otherwise switch to the new data.

src/data.py

@@ -1,7 +1,7 @@
 import io
 from pathlib import Path
 import sys
-import load_yaml
+from src import load_yaml
 from pandas import DataFrame, read_csv
 
 DEFAULT_YAML_PATH = Path("02-data/processed/relevant")