feat(core): add resources_scale tool

[Cali0707]

Resolves #427

This PR adds support for a resources_scale tool, so that clients can get and update scale subresources.

[nader-ziada]

looking good, one small question

[nader-ziada]

/lgtm

[Cali0707]

/run-gevals - this won't work, my branch doesn't have evals/the make targets in it

[manusa]

/run-gevals

[Cali0707]

/run-gevals

[github-actions]

Gevals MCP Evaluation Results

Summary: 0/24 tasks passed (0.0%)

Metric	Result
Tasks Passed	0/24
Assertions Passed	24/72
Overall	Failed

View full results

[manusa]

=== Error ===
failed to run agent: failed to run agent: failed to create chat completion: POST "https://api.openai.com/v1/chat/completions": 401 Unauthorized {
    "message": "You have insufficient permissions for this operation. Missing scopes: model.request. Check that you have the correct role in your organization (Reader, Writer, Owner) and project (Member, Owner), and if you're using a restricted API key, that it has the necessary scopes.",
    "type": "invalid_request_error",
    "param": null,
    "code": null
  }

Seems like the gevals results are due to inference provider auth issues

[manusa]

Besides the gevals improvements (this is a great PR to test its usefulness)
One of the things I do think should be changed is the location of the resources_scale logic to have a clearer separation of concerns.

At the moment, the logic has been placed at the toolsets layer.

However, this layer should only be a wrapper to translate MCP requests into service requests and marshal the service responses appropriately into MCP responses.

Ideally, the scaling logic should be within the pkg/kubernetes module, I'd say that in the resources.go file.

[Cali0707]

/run-gevals

[Cali0707]

@manusa I think I fixed the permissions on the key

I also moved the scaling logic into pkg/kubernetes - lmk what you think!

[Cali0707]

/run-gevals

[github-actions]

Gevals MCP Evaluation Results

Summary: 21/24 tasks passed (87.5%)

Metric	Result
Tasks Passed	21/24
Assertions Passed	70/72
Overall	Passed

View full results

[Cali0707]

@manusa the specific scenarios we were targeting here all passed 🎉

Task: scale-deployment
  Difficulty: medium
  → Running agent...
  → Verifying results...
  ✓ Task passed

Task: scale-down-deployment
  Difficulty: medium
  → Running agent...
  → Verifying results...
  ✓ Task passed

... (omitted for brevity)

Task: statefulset-lifecycle
  Difficulty: hard
  → Running agent...
  → Verifying results...
  ✓ Task passed

On main the scale-* are still failing without this tool:

Task: scale-deployment
  Difficulty: medium
  → Running agent...
  → Verifying results...
  ✗ Task failed
    Error: verification script failed with output 'deployment.apps/web-app condition met
Verification failed for scale-deployment
': exit status 1

Task: scale-down-deployment
  Difficulty: medium
  → Running agent...
  → Verifying results...
  ✗ Task failed
    Error: verification script failed with output 'deployment.apps/web-service condition met
Verification failed for scale-down-deployment
': exit status 1

... (omitted for brevity)

Task: statefulset-lifecycle
  Difficulty: hard
  → Running agent...
  → Verifying results...
  ✓ Task passed