Updating the `node-images.md` doc about cluster-api-provider-hetzner

[janiskemper]

@guettli can you give your input on this matter? Whatever format you can is appreciated!

[guettli]

@Sayanta66

Need live demos of the process that can be recorded instantly or recorded videos of the demos.

I would avoid that now. First it is a lot of work, and second a video can't be easily updated in the long run. Changing text is much easier.

A video makes sense to present a general overview of what caph is, but these details are better handled with written docs (my point of view).

[guettli]

@Sayanta66 you need to differentiate between hcloud machine images and bare-metal. The process is very different.

For bare-metal you can use any tool you like, and finally you need to create a tgz.

See: https://github.com/syself/cluster-stacks/blob/main/providers/hetzner/apalla/1-27/node-image/scripts/bm-export.sh

You can install the machine image via an oci-registry: there are already docs about that: https://github.com/syself/cluster-api-provider-hetzner/blob/main/docs/reference/hetzner-bare-metal-machine-template.md?plain=1#L100

One note: I would tell the user, that the most simple solution is to use preKubeadmCommands this is much easier than creating a custom image.