Consistent empty multi-value dimension behavior for groupBy / topN

Currently, groupBy and topN treat empty multi-value dimensions differently. groupBy treats them like nulls (i.e. empty strings) and topN ignores them (they don't contribute to the results at all).

Consider a dataset called `tweets` that has tweets, with a multi-value dimension `hashtags` listing the hashtags found in a tweet. It could be empty (a tweet with no hashtags) or it could have potentially multiple values, for a tweet like this one: https://twitter.com/sullcrom/status/1006208351095676929.

```sql
select hashtags, count(*)
from tweets
where floor(__time to hour) = timestamp '2018-06-22 00:00:00'
group by 1
order by 2 desc
limit 5
```

The groupBy engine returns:

```
hashtags,EXPR$1
,118719
KCAMexico,646
NBADraft,518
เป๊กผลิตโชค,249
BrasilGanha,199
```

And topN returns:

```
hashtags,EXPR$1
KCAMexico,646
NBADraft,518
เป๊กผลิตโชค,249
BrasilGanha,199
GOT7,159
```

The Druid docs don't seem to specify which behavior is correct for grouping: http://druid.io/docs/latest/querying/multi-value-dimensions.html. We should define one of these behaviors as correct and make the two engines consistent. I think aesthetically I prefer how topN works — why should empty lists be treated like a list containing a null? — but I am not sure how to reconcile that with the possibility that a groupBy could group by a multi-value dimension _and_ a single-value dimension. What if you group by `hashtags` and `username`, and some user never uses any hashtags? Should the fact that `hashtags` is empty make the rows implode, and that user would never show up in the results?

Start by reading the linked Druid multi-value dimensions documentation and comparing the groupBy and topN behavior described in the SQL example. Decide how empty multi-value dimensions should behave, including when grouped with a single-value dimension, then define the expected consistent results for both engines and verify them with focused queries or tests.

Written by the indexing model from the issue text.