Track and report managed/native memory footprint

[dblock]

Is your feature request related to a problem? Please describe.

In opensearch-project/OpenSearch#9403 a bug was fixed that caused native memory to increase in certain circumstances.

Describe the solution you'd like

Track and compare both JVM and native memory across benchmark runs (all the usual suspects of min/max/etc.).

Additional context

This is also one of the asks in #103.

[jmazanec15]

For native memory, Im guessing this could be supported by custom telemetry device for your purposes or maybe node stats. Maybe something like https://github.com/opensearch-project/opensearch-benchmark/blob/main/osbenchmark/telemetry.py#L628. Is the native memory youre interested in monitoring file-backed or anonymous? total virtual physical memory being used I believe can be gotten from node stats.