Ensure that JSON XBRL matches specification precisely on both input and output

[joelebwf]

This will require further conversations with XBRL and is being deferred to 1.1.0. One aspect of this is whether input validation is strict (i.e. must match exactly) or if approximate values could be handled. This also will be discussed in 1.1.0.

Notes: This appears to be the best approximation of the XBRL JSON specification. Further clarification is needed. For the time being just apply the correct values for output and determine the long term course on 1.1.0.

Best guess interpretation of XBRL JSON:
"key": false - correct
"key": "false" - incorrect
"key": "False" - incorrect
"key": "0" - incorrect
"key": 0 - incorrect

"key": true - correct
"key": "true" - incorrect
"key": "True" - incorrect
"key": "1" - incorrect
"key": 1 - incorrect

"key": null - correct
"key": "null" - incorrect
"key": "Null" - incorrect
"key": "None" - incorrect