[FEATURE] Flush benchmarking results after each test

[hughperkins]

Description

Flush benchmarking results after each test

Related Issue

Resolves Genesis-Embodied-AI/Genesis#
Resolves #1332

Motivation and Context

Having to wait until all benchmarking tests have run in order to see the results

How Has This Been / Can This Be Tested?

Running the benchmark script

pytest -svx tests/test_rigid_benchmarks.py -m benchmarks -sv

Screenshots (if appropriate):

Checklist:

• I read the CONTRIBUTING document.
• I followed the Submitting Code Changes section of CONTRIBUTING document.
• I tagged the title correctly (including BUG FIX/FEATURE/MISC/BREAKING)
• I updated the documentation accordingly or no change is needed.
• I tested my changes and added instructions on how to test it for reviewers.

• I have added tests to cover my changes.
• All new and existing tests passed.

[duburcqa]

Are you sure it changes anything ? I’m surprised. If you have a single GPU, you could just do:
pytest --print -x -n 0 -m 'benchmarks' ./tests

[hughperkins]

It definitely changes the output to that file. Because I tried it multiple times without, and had to wait 30 minutes to find out stuff wasnt working etc. And with, then eaach test outputs immediately.