Integrate network policy management

For security resons it would be great to integrate network policies to limit network connections only between allowed services.

Let's see an example:
let's assume we have an application composed of
- a **front-end web application** (ex. Nuxt.js or Nodejs server with Express) that will give a web representation of the data fetched by the back-end API
- a **back-end API** (ex. a Go application) that will read and interact with the data
- a **local Redis** service
- a **shared Mysql** service (shared between different projects as Mysql scales better if you have one global instance instead of multiple separated ones)

Now the **front-end App** should be contacted only by the ingress and can only contact the **back-end API**.
The **back-end API** should be contacted only by the **front-end App** but can connect to the **local Redis** and **shared Mysql** instances.
The **shared Mysql** instances may be contacted by multiple services in various projects while the other services are only available inside the pipeline/stage and cannot be accessed anywhere else (except from the web ingress).

Creating egress/ingress rules allows limiting access from ingoing and outgoing traffic where not allowed; let's say for example that the **front-end web application** get compromised:
in a normal kubernete system the attacker may try to reach other services in the cluster (ex: a not so well secured internal services that may contain critical data, only secured by "not opening any port to the public" ); by applying those rules the attacher may only reach the **back-end API** and cannot reach in any way non essential services.

To convert this into an UI/UX feature i propose to block by default (maybe adding also a checkbox to disable this feature?) all the traffic that is not "linked" to the service as an add-on or as a "linked service" from another pipeline (in this example the global Mysql service, other services may include Mail services, etc.).