Currently reporters always produce data row by row.
Output buffer also receives data in that form and saves each row, making sure that every row is available as soon as possible for consumption.
This is only good for live reporting (documentation reporter) but is of no value for post-run reports like coverage reporting.

The row-by-row for those reports is causing significant load and performance impact on overall runtime of tests.

We should have output buffer capability for writing data in bulks.
That can be achieved by passing lines to process as CLOB or by passing a list of lines to process.