FluxCD: Application update rollback from armada does not remove fluxcd release info

[Lucas]

Brief Description
-----------------
If an system application-update is triggered updating an armada app to a fluxcd app (preceded by a helm release migration) and update fails, application framework will try to perform a recover, such recover will fail as fluxcd uses helm3 and armada helm2, this will create resources both in helm2 and helm3 leaving the system in a inconsistent state

Severity
--------

Major: updated application that failed must be deleted

Steps to Reproduce
------------------
1. apply armada app
2. migrate helm release from app
3. system-application update <fluxcd_tarball>
4. system-application abort app

Expected Behavior
------------------
recover does not start when coming from armada to fluxcd

Actual Behavior
----------------
recover starts and breaks the app metadata
Reproducibility
---------------
reproducible

System Configuration
--------------------
any

Branch/Pull Time/Commit
-----------------------
master

Last Pass
---------
na

Timestamp/Logs
--------------
cert-manager xxx cert-manager-fluxcd-manifests fluxcd-manifests applied completed
nginx-ingress-controller xxx nginx-ingress-controller-manifest nginx_ingress_con apply-failed Application
                                                                      troller_manifest. update from
                                                                      yaml version xxx
                                                                                                      to version yyy aborted.
                                                                                                      Application
                                                                                                      recover to
                                                                                                      version xxx
                                                                                                      aborted. Please
                                                                                                      check logs for
                                                                                                      details.

Test Activity
-------------
Upgrade testing

Workaround
----------
delete app

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.opendev.org/c/starlingx/config/+/848133

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/c/starlingx/config/+/848133
Committed: https://opendev.org/starlingx/config/commit/5025e1ce710f626012e5e51f492e637483c4a245
Submitter: "Zuul (22348)"
Branch: master

commit 5025e1ce710f626012e5e51f492e637483c4a245
Author: Lucas Cavalcante <email address hidden>
Date: Wed Jun 29 10:48:31 2022 -0300

    Skip Recover From Armada to FluxCD apps

    If an system application-update is triggered updating an armada app
    to a fluxcd app (preceded by a helm release migration) and
    update fails, the application framework will try to perform a recover.

    Recover will fail as fluxcd uses helm3 and armada helm2. This will
    create resources both in helm2 and helm3 leaving the app
    in a inconsistent state.

    To prevent that from happening recover is skipped if to_app and from_app
    use different chart managers.

    TEST PLAN:
    PASS: recover skipped after update from armada to fluxcd without
    migrating helmrelease

    Closes-bug: 1980242
    Signed-off-by: Lucas Cavalcante <email address hidden>
    Change-Id: I9061b75f443730e973b79cc93e955069951113ff