Add DataFrame.es_query() to modify Elasticsearch queries directly

[sethmlarson]

Follows the form of prefixing anything Elasticsearch-related with es_.

Let me know if you think this should be implemented differently, this seemed like the straightforward way to me and seemed to work for a handful of query types and combinations. :)

Allows users to specify queries like so:

>>> import eland as ed
>>> ed.DataFrame("localhost", "ecommerce")
>>> df.es_query({
  "bool": {
    "filter": {
      "geo_distance": {
        "distance": "1km",
        "geoip.location": [55.3, 25.3]
      }
    }
  }
})[["category", "geoip.location"]].head(5)
                                                  category              geoip.location
1                                       [Women's Clothing]  {'lon': 55.3, 'lat': 25.3}
9                        [Women's Shoes, Women's Clothing]  {'lon': 55.3, 'lat': 25.3}
10  [Women's Accessories, Women's Clothing, Women's Shoes]  {'lon': 55.3, 'lat': 25.3}
22                       [Women's Shoes, Women's Clothing]  {'lon': 55.3, 'lat': 25.3}
30                                        [Men's Clothing]  {'lon': 55.3, 'lat': 25.3}

Closes #143.

[stevedodson]

Nice PR! Only comment is to check doctests

[stevedodson]

LGTM