Clean CSV table of figures and captions from public Plazi endpoint

This table was generated by querying Plazi's public SPARQL endpoint at https://treatment.ld.plazi.org/sparql, using the query figures-and-captions.sparql.

The query was made using the following curl command:
```
curl -o figures-and-captions.plazi.csv -H "Accept: text/csv" --get --data-urlencode [email protected] https://treatment.ld.plazi.org/sparql
```

.gitattributes

@@ -54,3 +54,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.jpeg filter=lfs diff=lfs merge=lfs -text
 *.webp filter=lfs diff=lfs merge=lfs -text
 figures-and-captions.tsv filter=lfs diff=lfs merge=lfs -text
+figures-and-captions.plazi.csv filter=lfs diff=lfs merge=lfs -text

figures-and-captions.plazi.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16e8488f7b633fa825c55d9d00338a1e2e0916922964ba30bbb8430867e48cc3
+size 124925673