Fix type inconsistencies in 6.x docs examples

[cjcenizal]

For example, the Index Templates 6.0 docs have an example that suggests using a type:

PUT _template/template_1
{
  "index_patterns": ["te*", "bar*"],
  "settings": {
    "number_of_shards": 1
  },
  "mappings": {
    "type1": {
      "_source": {
        "enabled": false
      },
      "properties": {
        "host_name": {
          "type": "keyword"
        },
        "created_at": {
          "type": "date",
          "format": "EEE MMM dd HH:mm:ss Z YYYY"
        }
      }
    }
  }
}

However, the Index Templates 6.5 docs suggest using the _doc type.

PUT _template/template_1
{
  "index_patterns": ["te*", "bar*"],
  "settings": {
    "number_of_shards": 1
  },
  "mappings": {
    "_doc": {
      "_source": {
        "enabled": false
      },
      "properties": {
        "host_name": {
          "type": "keyword"
        },
        "created_at": {
          "type": "date",
          "format": "EEE MMM dd HH:mm:ss Z YYYY"
        }
      }
    }
  }
}

I suggest we reconcile these types of inconsistencies throughout 6.x so that users have an easier time understanding and accommodating the types deprecation change.

[elasticmachine]

Pinging @elastic/es-docs

[DaveCTurner]

Thanks @cjcenizal, I think the particular discrepancy you highlight is somewhat deliberate, because _doc wasn't a legal type name until #27816 (i.e. 6.2.0), so we can't use _doc in the 6.0 reference documentation. That PR adjusts quite a lot of the docs, and there were a few followup PRs to deal with things it missed which did not all get backported all the way to 6.2. The case you highlight was fixed in #29667 (i.e. 6.4.0). Would you be interested in back porting that to 6.3 and 6.2?

As a user, the best bet is to upgrade to 6.8 (which is compatible with earlier 6.x releases) and then use the 6.8 docs.

[elasticmachine]

Pinging @elastic/es-search

[cbuescher]

+1 to what @DaveCTurner said. Using "_doc" as the "dummy" type in later 6.x documentation is necessary, but at the same time we cannot use it going backwards. I understand the irritation but the docs are in sync with what each of the minor versions requires in this case I think.
@cjcenizal does this explanation make sense to you and would you mind if we close this issue?

[cjcenizal]

@cbuescher Yes, this makes sense. Thanks for the explanation. I'll close this issue and backport #29667 to 6.3 and 6.2 per @DaveCTurner's request.