fix: Populate SQL_CONNECT_ATTRIBUTES in admintools and schema job

[adigiorgi-clickup]

What was changed

When server.persistence.config.*.sql.connectAttributes is set, also populate the SQL_CONNECT_ATTRIBUTES env var in the temporal-schema job. According to https://github.com/temporalio/temporal/blob/main/tools/sql/handler.go#L157, the SQL_CONNECT_ATTRIBUTE env var must be in this format: key1=value1&key2=value.

Why?

If users want to use a different schema than public, they need to set the connectAttribute value to something like search_path: temporal. With this PR, the temporal-schema job will know about it and create tables in the right schema.

Checklist

1. Closes [Feature Request] Ability to specify SQL_CONNECT_ATTRIBUTES for temporal-schema job #561

2. How was this tested:

   1. helm lint charts/temporal
   2. Add connectAttributes to charts/temporal/values/values.postgresql.yaml:

   connectAttributes:
     search_path: temporal
     foo_bar: baz

   3. helm template my-release charts/temporal -f charts/temporal/values/values.postgresql.yaml
   4. Check the resulting yaml contains the SQL_CONNECT_ATTRIBUTES env var:

      - name: SQL_CONNECT_ATTRIBUTES
        value: "foo_bar=baz&search_path=temporal"

   5. Run step iii against main and against my branch, checking the resulting yamls are the same.

[CLAassistant]

All committers have signed the CLA.