Rename Neptune operator db_cluster_id argument to cluster_id

[mitre88]

Part of the template-field validation burn-down tracked in #70296.

NeptuneStartDbClusterOperator and NeptuneStopDbClusterOperator declare cluster_id in template_fields but expose the value as the db_cluster_id constructor argument, assigning self.cluster_id = db_cluster_id in __init__ — which the validate-operators-init check flags. Per review feedback, this renames the argument to cluster_id (rather than renaming the attribute as in the first iteration of this PR), so the templated attribute and template_fields entry stay exactly as they are on main and existing Dags templating the attribute are unaffected.

Breaking change: Dags passing db_cluster_id= to these two operators must switch to cluster_id=, and now fail loudly with a TypeError at construction time. A note was added to the provider changelog. The db_cluster_id keys in the operators' XCom return values and in the Neptune trigger/hook APIs are unchanged.

Both classes are removed from the exemption list and the validate-operators-init check passes locally, as do the unit tests.

---

Was generative AI tooling used to co-author this PR?

• Yes — Claude Code (Fable 5)

Generated-by: Claude Code (Fable 5) following the guidelines

[uranusjr]

Since this is breaking either way, maybe it’s easier to rename the argument instead?

[mitre88]

Good point — reworked the PR to rename the argument instead: the constructor now takes cluster_id, the templated cluster_id attribute and template_fields stay exactly as on main, and a wrong keyword fails loudly with a TypeError. Updated the unit tests, the system test example, and added a breaking-change note to the provider changelog. PR title and description updated accordingly.

---

Drafted-by: Claude Code (Fable 5) (no human review before posting)

[vincbeck]

This can be treated as a bug fix, I dont think we need a major version bump for this one

[o-nikolas]

To me it feels more correct to change the template field names, not the argument itself in the constructor. But it I wont die on that hill if everyone else things otherwise

[eladkal]

To me it feels more correct to change the template field names, not the argument itself in the constructor. But it I wont die on that hill if everyone else things otherwise

I believe this is more AWS decision how you want to play this. Probably the merge click is waiting for the team to decide if you are happy with it as is or would you like to ask author to change approach.

[vincbeck]

To me it feels more correct to change the template field names, not the argument itself in the constructor. But it I wont die on that hill if everyone else things otherwise

I agree with Niko, can we change the template field instead?