Consider continuing imports without metadata extraction if the GROBID service is unavailable
When a request to compile/document
includes the Pressroom-Enrich-Document-Metadata
header, the document is passed through GROBID to extract metadata.
If the GROBID service happens to be unavailable for some reason, it might be better to continue with the import as if the Pressroom-Enrich-Document-Metadata
header wasn't set, rather than returning an error response, as this is what the client's going to do anyway if it retries.