Log machine info in bencher workflow

[Copilot]

Benchmark runs land on whatever the 1ES pool hands out, with no record in the logs of which machine actually ran. Add a "Machine info" step to each benchmark job in the scheduled bencher workflow so results can be correlated with hardware.

• .github/workflows/bencher.yml — new step in benchmark_virtual and benchmark_snp, run right after checkout.

Each step runs:

uname -a
nproc
cat /proc/cpuinfo
cat /proc/meminfo

Beyond the requested cat /proc/cpuinfo, nproc reflects cores actually visible to the container (cgroup-aware, unlike /proc/cpuinfo), /proc/meminfo captures available RAM, and uname -a pins the kernel — all of which materially affect benchmark numbers.

Possible follow-ups, not done here to keep the change minimal and dependency-free: lscpu (cleaner summary, requires tdnf install util-linux in the Azure Linux container), dmidecode (needs privileged container), or emitting the info as a bencher testbed attribute rather than just to the log.

[maxtropets]

@copilot remove it from AB bencher yml

[Copilot]

@copilot remove it from AB bencher yml

Removed in cda8e2f — bencher-ab.yml is back to its original state, the Machine info step now only lives in bencher.yml.

[Copilot]

Pull request overview

Adds a lightweight “Machine info” logging step to the scheduled benchmark workflow so benchmark results can be correlated with the underlying runner hardware/kernel characteristics.

Changes:

• Add a Machine info step to benchmark_virtual to log kernel, CPU, and memory details immediately after checkout.
• Add the same Machine info step to benchmark_snp immediately after checkout.