Add auxiliary table to record results of regular queries (e.g. run as part of daily ingestion)

[maxalbert]

As part of the daily ingestion process one typically wants to run some initial quality checks or calculate informative metrics. Simple examples: how many call events happened on that date? Are there any cell_id values in this date's CDR data which are not present in the infrastructure.cells table (and if so how many)? Etc.

In the longer run this will be supported by a set of supplementary tables (to be added as part of #640) that record the results of standard queries (or standard pre-aggregations) which are run as part of the ingestion process.

In the immediate short term it will be useful to have a single auxiliary table where the results of semi-ad-hoc queries can be recorded in a slightly less structured and more informal way but which still allows quick querying of results. Suggestion for the structure of this table:

|   cdr_date | cdr_type | type_of_query_or_check |  outcome | optional_comment_or_description |
|------------+----------+------------------------+----------+---------------------------------|
| 2019-01-01 | calls    | num_total_calls        | 42000000 | NULL                            |
| 2019-01-01 | calls    | num_missing_cell_ids   |     4242 | needs investigation             |

Such a table should be added to flowdb, e.g. as etl.daily_queries (any suggestions for a better name?).