Dashboard messages should be more accurate when self-signed cert issue happens

[sleshchenko]

Describe the bug

When Che is deployed with self-signed certificate and the corresponding CA is not imported to browser - the only way to understand why Che fails to load in browser - use Developer Tools and analyze failed requests.
So, Dashboard messages should be more accurate when self-signed cert issue happens

Che version

• latest
• nightly
• other: please specify

Steps to reproduce

1. Deploy Che with self-signed certs
2. Open Che in a browser without CA imported
3. Add Che host as trusted.
4. Check error message:
   Actual: The authorization token is missing

Screenshot

Expected: Something like Keycloak host is signed with self-signed certificate. Add it to trusted hosts or import the corresponding CA into browser

Expected behavior

Error messages clearly describe an issue and ways how to fix it.

Runtime

• kubernetes (include output of kubectl version)
• Openshift (include output of oc version)
• minikube (include output of minikube version and kubectl version)
• minishift (include output of minishift version and oc version)
• docker-desktop + K8S (include output of docker version and kubectl version)
• other: (please specify)

[sleshchenko]

I believe it's really important since we have plans to enable TLS by default, and even without it I received a lot of complain that people do not understand what is wrong with their installation and consider such error as totally non-working Che, but in fact but more step is needed from user side and it's just not obvious.
cc @slemeur @l0rd

[sleshchenko]

The PR that fixes the case with Authorization token is missed is merged