feature request: single first outlier causes mean to be useless for comparison

[nolistic]

as a business user, I want to easily compare metrics across loadtest reports.

However, miliseconds matter in the world of response time.

Outliers cause me to need more robust reporting than is offered by loadtest when comparing platforms.

when separated across threads, this first outlier makes sense

but that doesn't mean I want it in my comparison report.

[jnolis]

It's easy to change the mean response time to median (in fact that is what the chart uses for the horizontal gray line), but should we do something in the first plot to remove that outlier? It makes the rest of the chart useless, but could be important.