Adding DataFrame.Dtypes support

[Niharikadutta]

This PR adds support for dtypes and addresses part of #621

Scala side:
Dataframe.dtypes => returns Array[(String, String)]

dtypes when called on a DataFrame, returns all column names and their data types as an array of tuples. For example, for a dataframe df with the following schema:
root
|-- age: long (nullable = true)
|-- name: string (nullable = true)

Calling dtypes:

val result = df.dtypes
result: Array[(String, String)] = Array((age,LongType), (name,StringType))

C# side:
Dataframe.DTypes() => returns IEnumerable<Tuple<string, string>>

Calling DTypes():

IEnumerable<Tuple<string, string>> result = df.DTypes();

[suhsteve]

@Niharikadutta I think you made this a bit more complicated than it needed to be. How about we do something like this instead?

public IEnumerable<Tuple<string, string>> DTypes() =>
            Schema().Fields.Select(
                f => new Tuple<string, string>(f.Name, f.DataType.SimpleString));

[Niharikadutta]

@suhsteve Good point, I didn't think about just exposing the functionality instead of calling spark's dtypes. This also returns C#'s DataTypes like integer and string instead of Spark's IntegerType and StringType, which is good I think?

[imback82]

Yea this should be fine. Please check pyspark implementation: https://github.com/apache/spark/blob/master/python/pyspark/sql/dataframe.py#L1010

[imback82]

LGTM except minor comments

[imback82]

Can you remove .ionide/symbolCache.db from this PR?

[suhsteve]

LGTM. Thanks @Niharikadutta !

[suhsteve]

Can you update your description to reflect the updated changes ?

[imback82]

LGTM, thanks @Niharikadutta!