e2e: enforce virtual machine cache topology levels

[askervin]

On some host platforms, Qemu reports every virtual machine CPU hyperthread having its own L1d, L1i and L2 caches. Force virtual CPU cache topology where these caches are shared between thread siblings.

[askervin]

This does not seem to be enough, problem persists (and gets even worse, see L3...) at least on one host system:

[root@n4c16-fedora-43-containerd ~]# grep . /sys/devices/system/cpu/cpu0/cache/index*/*_list
/sys/devices/system/cpu/cpu0/cache/index0/shared_cpu_list:0
/sys/devices/system/cpu/cpu0/cache/index1/shared_cpu_list:0
/sys/devices/system/cpu/cpu0/cache/index2/shared_cpu_list:0
/sys/devices/system/cpu/cpu0/cache/index3/shared_cpu_list:0-3

[root@n4c16-fedora-43-containerd ~]# grep . /sys/devices/system/cpu/cpu0/topology/{thread_siblings_list,package_cpus_list}
/sys/devices/system/cpu/cpu0/topology/thread_siblings_list:0-1
/sys/devices/system/cpu/cpu0/topology/package_cpus_list:0-7

--> keeping the PR as Draft.