SageMaker Studio Pipeline UI shows generic error instead of actual permission error

### Product Version

- [ ] Amazon SageMaker Studio Classic
- [x] Amazon SageMaker Studio
- [ ] Issue is not related to SageMaker Studio

### Issue Description

# Problem
When a user lacks IAM permission (e.g., `sagemaker:DescribePipelineExecution`), the Pipeline UI shows generic, unhelpful errors:
- "An error occurred when loading pipeline Pipeline-XXXXX"
- "ARN undefined"
- Empty canvas

# Expected Behavior
The UI should display the actual error message returned by the API, such as: "AccessDeniedExeception": User is not authorized to perform sagemaker:DescribePipelineExecution"

# Actual Behaviror
Generic error message that doesn't indicate the root cause. Users have to switch to the Python SDK to discover the real error in the traceback.

# Impact
- Wasted debugging time
- Users may think the UI is broken when it's actually a permission issue
- Poor user experience

# Steps to Reproduce
1. Create an "Amazon SageMaker AI Domain" with a role missing DescribePipelineExecution permission or AmazonSageMakerFullAccess permission.
2. Create and execute a pipeline via the visual editor.
3. A message at the bottom will pop up "Successfully created pipeline execution `View pipeline execution`".
4. Click on this `View pipeline Execution` button. It will show "arn undefined error"
5. Go back to pipelines and click on the pipeline. It will show "an error occurred when loading the pipeline Pipeline-XXXXX"

# Environment
- SageMaker Studio (new version, not Classic)
- Region: us-east-2

I have attached the screenshots of the error messages.

### Expected Behavior

The UI should display the actual error message returned by the API, such as: "AccessDeniedExeception": User is not authorized to perform sagemaker:DescribePipelineExecution"

### Observed Behavior

Generic error message that doesn't indicate the root cause. Users have to switch to the Python SDK to discover the real error in the traceback.

### Product Category

Pipelines

### Feedback Category

User Interface

### Other Details

_No response_

Start by reproducing the failure in Amazon SageMaker Studio in us-east-2 with a role lacking sagemaker:DescribePipelineExecution, then compare the Pipeline UI response with the Python SDK traceback. Done means the UI shows the API's permission error, including the denied action, instead of a generic loading error, undefined ARN, or empty canvas.

Written by the indexing model from the issue text.