running "busybox gzip -t" is not so useful?

The current postmarketos-mkinitfs code attempts to extract the generated initramfs with busybox gzip. This was my suggestion, so I'm to blame 😅

I think it would be better if we removed it again:

• if I understand it right, it runs for regular initramfs and for initramfs-extra now. But busybox will only extract the initramfs-extra, the regular initramfs is extracted by the kernel (in almost all cases, except the weird i9100 case)
• if we wanted to fully test that busybox can extract this, we would also need to extract the cpio archive (which means we can't use the -t test command with gzip)
• in general, busybox extraction code should be pretty solid. it should not be necessary to verify that extraction works every time we generate an initramfs and waste time doing that, on every user's device. I didn't realize how much time it would take when I proposed this.
• even if there should be a bug introduced into busybox extraction code, it will be in alpine edge first. we will catch it there before it hits the stable releases of postmarketOS.

So yeah sorry for requesting this feature first and then asking to remove it again. But I think it would be better if we get rid of it.

If we want to do some test of this sort, it could be useful to create a CI test that generates an initramfs and then tries to extract the generated files. Then it only runs when postmarketos-mkinitfs changes, not all the time for each user when they build a new initramfs.