Facets-overview

* leverage HTML <link rel="import"> we imported facets-jupyter.html, as result, we don't need to keep
all the styles and CSS for javascripts code. We can need index.html
* make copy of data for easy demo presentation ( although the same file has multiple copies)
* removed SQLContextFactory.scala as its no longer needed
* Jupyter notebook also updated to reference <facets-jupyter.html> from google facets github instead of local copy

it to String instead.

* fix field did not correctly specify data type after converted to numeric data frame
* fix field values are all nulls, which code can't determine the category of the data type
* Note the date field converted to numerical field is based on the facets_overviews' covert np.datetime64 to INT
but now, I am not sure this is right approach to treat a date field as Integer field.
as if the data is originally from a CSV file with Date String, it will be treated the as STRING,
but if the date is selected from Database where the Date String is stored in Date/Timestamp field, then current code
will convert it to numeric field and calculate of stats of the long value. Its not very helpful. But I might fix it in next PR.
* add more tests