docs: refactor resources, add gpu, kong

[tarrantro]

Goals:

• Add document for JCloud gateway

• Add document for GPU support

[JoanFM]

@deepankarm @JohannesMessner ,

can you please review the wording?

[JoanFM]

Also make sure to clarify the difference between core Gateway and this Kong Gateway

[JohannesMessner]

There are a few things that are not quite clear to me (probably because I don't know much about these technologies, but I imagine some of our users might be in a similar position):

• The two supported Gateways are ALB and Kong, right? Then I wouldn't put theme in a note note-box, rather put them in a listing or just as separate sub-sections, depending on how much additional details are presented for each of them
• I don't know what layer 7 and layer 4 load balancing mean, what are their implications for me as a user?
• Apart from that, certs seem to be the main difference between the two. When should I choose what on that dimension?
• It is suggested to use Kong, but why? What are the advantages I get as a user?

Again, some of these questions might be super trivial, but it would be cool if you could provide a bit more context around them!

[JoanFM]

We need to explain that this has nothing to do with our Gateway inside Jina

[JohannesMessner]

The added explanation about level 7 and level 4 is great for me!

I still have a few suggestions (because I am picky) and also please don't use backticks for highlighting, we decided to only use them to render short code snippets (classes/methods/yaml optons etc.).

[deepankarm]

@tarrantro I'll work on editing a few sections in this PR. Making it a draft now.

[tarrantro]

add GPU document as well

[delgermurun]

Don't forget to remove "JCloud doesn’t support GPUs yet." from https://docs.jina.ai/fundamentals/jcloud/faq/ 😄

[JohannesMessner]

I like the description of GPU support. Just two points:

• Is a provisioned GPU usable for the entire Flow? Or can it be assigned on a per-Executor basis? This is not so clear to me
• If dedicated GPU is the default way of doing it, that should probably come before shared GPU

[github-actions]

📝 Docs are deployed on https://docs-add-kong--jina-docs.netlify.app 🎉

Addressed