helm-controller sharding not working in 2.2.1

[MaxRink]

Describe the bug

We have two instances of the helm operator in each cluster, one for platform components, another one for everything else.
With the latest release both controllers try to reconcile all HRs, which leads to 'interesting' behaviour
e.g.

  Normal   UpgradeSucceeded     26m                helm-controller  Helm upgrade succeeded for release schiff-system/schiff-network.v2 with chart schiff-network@0.1.0
  Warning  UpgradeFailed        26m                helm-controller  Helm upgrade failed for release schiff-system/schiff-network with chart schiff-network@0.1.0: release: already exists

Steps to reproduce

Flux 1 has:
--watch-label-selector=sharding.fluxcd.io/key=platform
Flux 2 has:
--watch-label-selector=sharding.fluxcd.io/key notin (platform)
i also tried
--watch-label-selector=!sharding.fluxcd.io/key
just to make sure (which is in the current docs)
Both worked in the past and there was no change in any of the changelogs

The HRs either have

  labels:
    sharding.fluxcd.io/key: platform

or no label at all

Expected behavior

The sharding works as expected.

Screenshots and recordings

No response

OS / Distro

N/A

Flux version

v2.2.1

Flux check

N/A

Git provider

No response

Container Registry provider

No response

Additional context

No response

Code of Conduct

• I agree to follow this project's Code of Conduct

[hiddeco]

Sorry you had to run into this Max. We addressed this in the above PR, and CI is currently building v0.37.2 which should be available in a couple minutes. As soon as that's available, I will do an emergency release of the CLI as well.

But in the meantime, updating just the controller should be sufficient.

[hiddeco]

Now available as https://github.com/fluxcd/flux2/releases/tag/v2.2.2