Add fio disk performance test CS workload

With new integrity file-system changes, we need a way to ensure disk
performance is consistent across releases of Confidential Space.
Followed instructions at:
https://cloud.google.com/compute/docs/disks/benchmarking-pd-performance#existing-disk
Separate change to automate this test.

launcher/image/testworkloads/diskperf/Dockerfile

@@ -0,0 +1,13 @@
+# From current directory:
+# gcloud builds submit --tag us-west1-docker.pkg.dev/confidential-space-images-dev/cs-integ-test-images/diskperf:latest --project confidential-space-images-dev
+# Example metadata to configure fio:
+# --metadata="^~^tee-image-reference=us-west1-docker.pkg.dev/confidential-space-images-dev/cs-integ-test-images/diskperf:latest~tee-restart-policy=Always~tee-cmd=[\"/jobs/default.fio\", \"--size=50G\", \"--numjobs=1\"]"~tee-container-log-redirect=true
+FROM alpine
+
+RUN apk add --no-cache fio
+COPY jobs /jobs
+
+LABEL "tee.launch_policy.log_redirect"="always"
+LABEL "tee.launch_policy.allow_cmd_override"="true"
+
+ENTRYPOINT ["fio"]

launcher/image/testworkloads/diskperf/jobs/default.fio

@@ -0,0 +1,39 @@
+; https://cloud.google.com/compute/docs/disks/benchmarking-pd-performance#existing-disk
+[global]
+time_based
+runtime=60s
+ramp_time=2s
+ioengine=libaio
+direct=1
+verify=0
+group_reporting=1
+
+[write_throughput]
+#numjobs=16
+bs=1M
+iodepth=64
+rw=write
+iodepth_batch_submit=64
+iodepth_batch_complete_max=64
+
+[write_iops]
+bs=4K
+iodepth=256
+rw=randwrite
+iodepth_batch_submit=256
+iodepth_batch_complete_max=256
+
+[read_throughput]
+#numjobs=16
+bs=1M
+iodepth=64
+rw=read
+iodepth_batch_submit=64
+iodepth_batch_complete_max=64
+
+[read_iops]
+bs=4K
+iodepth=256
+rw=randread
+iodepth_batch_submit=256
+iodepth_batch_complete_max=256