Consider writing 'endpoint' to SPARQL JSON

[ross-spencer]

The JSON could further extended to include the endpoint URL. Doing so provides further information to the caller of the JSON output, be that immediate consumption, or like in Siegfried's case, from a file on disk.

[ross-spencer]

Related to: ross-spencer/siegfried#2

[ross-spencer]

Example output might be:

{
  "endpoint": "http://wikibase.uri/proxy/wdqs/bigdata/namespace/wdq/sparql?",
  "head": {
    "vars": [
      "uri",
      "uriLabel",
      "puid",
      "extension",
      "mimetype",
      "encoding",
      "referenceLabel",
      "date",
      "relativity",
      "offset",
      "sig"
    ]
  },