FIX: adjust packet metadata url and apiVersion for new equinix api

[RiRa12621]

adjust packet metadata url and apiVersion for new equinix api

The equinix metadata API changed since they acquired packet, causing issues during boot.

How to use

• build
• run on equinix
  To test an out of stream change on equinix, you'll have to start a machine using ipxe boot, as equinix only allows stable and beta channels

Testing done

• Changelog entries added in the respective changelog/ directory (user-facing change, bug fix, security fix, update)
• Inspected CI output for image differences: /boot and /usr size, packages, list files for any missing binaries, kernel modules, config files, kernel modules, etc.
• ensure instance with image completely starts up on equinix

Notes

Since we're making heavy use of this, we're happy to help testing this out, by ie providing an instance on Equinix for testing

[RiRa12621]

related conversation from k8s-slack --> https://kubernetes.slack.com/archives/C03GQ8B5XNJ/p1723144757084059

[jepio]

Tested this:

# Before:
$ ./coreos-cloudinit --oem=packet
2024/08/19 09:19:32 Checking availability of "packet-metadata-service"
2024/08/19 09:19:32 Checking availability of "packet-metadata-service"
2024/08/19 09:19:32 Checking availability of "packet-metadata-service"
2024/08/19 09:19:33 Checking availability of "packet-metadata-service"
2024/08/19 09:19:33 Checking availability of "packet-metadata-service"
^C
# After:
$ ./coreos-cloudinit --oem=packet
2024/08/19 09:21:29 Checking availability of "packet-metadata-service"
2024/08/19 09:21:29 Fetching meta-data from datasource of type "packet-metadata-service"
2024/08/19 09:21:29 Fetching data from https://metadata.packet.net/metadata. Attempt #1
2024/08/19 09:21:30 Fetching user-data from datasource of type "packet-metadata-service"
2024/08/19 09:21:30 Fetching data from https://metadata.packet.net/userdata. Attempt #1
2024/08/19 09:21:31 Failed to set hostname: error setting hostname: exit status 1
2024/08/19 09:21:31 Authorized SSH keys for core user

Failures are due to running as non-root, so all looks good.