TT-12175 Provider selection behavior change in >v1.1.1

[sarnowski23]

Branch/Environment

• Tag: >v1.1.1
• Environment: Kubernetes

Describe the bug
After v1.1.1, attempting to access tib openid-connect provider fails with {"Status":"error","Error":"you must select a provider"}.

Reproduction steps
Steps to reproduce the behavior:
Profiles.json:

[{
  "ID": "keycloak",
  "OrgID": "<orgid>",
  "ActionType": "GenerateOrLoginUserProfile",
  "Type": "redirect",
  "ProviderName": "SocialProvider",
  "ProviderConfig": {
      "CallbackBaseURL": "https://tib.<domain>",
      "FailureRedirect": "https://tyk.<domain>/?fail=true",
      "UseProviders": [
      {
          "DiscoverURL": "https://keycloak.<domain>/auth/realms/<realm>/.well-known/openid-configuration",
          "Key": "<client_name>",
          "Name": "openid-connect",
          "Scopes": [
          "openid",
          "email"
          ],
          "Secret": "<secret>",
          "SkipUserInfoRequest": false
      }
      ]
  },
  "ReturnURL": "https://tyk.<domain>/tap",
  "SSOOnlyForRegisteredUsers": false
}]

Access url:
https://tib.<domain>/auth/keycloak/openid-connect

Actual behavior
I have a client configured in keycloak and accessing via the access url noted above works correctly (redirecting to keycloak and then back to tyk authenticated) on image version v1.1.1. On v1.2.1 and above (1.2.0 is just broken yielding exec format error and crashing on startup) this yields an error '{"Status":"error","Error":"you must select a provider"}'. My provider setup is as documented here.

Expected behavior
I'd expect provider selection to work as before, or documentation indicating the new pattern.

Additional context
Running in an EKS cluster. Tyk deployed with official helm charts: https://github.com/TykTechnologies/tyk-helm-chart/tree/master/tyk-pro v0.14.0, previously v0.13.2.