Support printing other types like scalars, arrays, etc.

Support printing other types like scalars, arrays, etc with the query API endpoint and the CLI.

Expectation:
When users use any of the following functions:

http.get()
findRecord()
getRecord
findColumn()
usage.limits
...and any other functions I'm missing.
They must construct a temporary table and map the output to it in order to output these other formats. In otherwords they must use either use array.from() and map() or follow the example here:
https://docs.influxdata.com/influxdb/v2.0/reference/flux/stdlib/experimental/http/get/#get-the-status-of-influxdb-oss

Reality:
The query runs, and no output is generated. The extension doesn't provide any errors when running these queries so the users are unaware of the need to search for these hacks. Related to influxdata/flux#4654 and influxdata/vsflux#234

Proposal
Provide support to print other data types.

Start by tracing the query API endpoint and CLI paths that handle results from http.get(), findRecord(), getRecord, findColumn(), and usage.limits. Compare how scalar and array results are currently handled versus table output, then identify the relevant tests or entry points. Done means these functions print non-table results directly without temporary-table workarounds and provide errors when output cannot be handled.

Written by the indexing model from the issue text.