auto_join with k8s as provider can auto apply to current namespace

Is your feature request related to a problem? Please describe.

Ideally, we expect use helm chart with model:

helm install ./vault-helm --namespace MY_NS

everytime, when we switch namespaces, expect MY_NS as only parameter we need to modify.

However, if set auto_join with k8s as provider, the config is like:

      storage "raft" {
          path = "/vault/data"
          retry_join {
            auto_join = "provider=k8s label_selector=\"app.kubernetes.io/name=vault,component=server\" namespace=\"MY_NS\" "
            auto_join_port = 8200
            leader_tls_servername = "vault"
            auto_join_scheme = "https"
            leader_ca_cert_file = "/vault/cert/vault.ca"
            leader_client_key_file = "/vault/cert/vault.key"
            leader_client_cert_file = "/vault/cert/vault.crt"
          }
        }

We have to manually define the namespace here if we expect Vault only discover PODs within same namespace.

the root cause is that go-discover need explicitly argus input.
https://github.com/hashicorp/go-discover/blob/master/provider/k8s/k8s_discover.go#L103

Describe the solution you'd like

Maybe something like:

let k8s_discover.go accept environment variable.

helm chart feeds .Release.Namespace to this environment variable with configuration flag.

Eg:
below this line:https://github.com/hashicorp/vault-helm/blob/main/values.yaml#L788

      # Set the setDiscoverNamespace to current deploy namespace
      setDiscoverNamespace: true

Describe alternatives you've considered
N/A

Additional context
N/A

Start by reading go-discover's k8s_discover.go around the referenced argument handling, then inspect vault-helm's values.yaml near the setDiscoverNamespace setting. Trace how the chart can expose .Release.Namespace to discovery without requiring a manually edited namespace. Done means auto_join discovers pods in the current namespace when enabled, while explicit namespace configuration remains usable.