@aws-cdk/aws-servicecatalog Ephemeral storage modification support

### Describe the feature

Currently, AWS CDK’s @aws-cdk/aws-servicecatalog module uses an internal Lambda function for asset copying during Service Catalog product deployment. This Lambda’s configuration (e.g., **ephemeral storage**) is not exposed for user customization.

This feature would allow developers to override or specify these Lambda **ephemeral storage** when deploying Service Catalog products.

Expected Enhancements:
1. Add properties (e.g., lambdaProps) to allow users to configure the internal Lambda's **Ephemeral storage** (similar to constructs like aws-lambda.Function OR aws-s3-deployment.BucketDeployment).
2. Ensure this configurations is exposed as optional parameters during servicecatalog.ProductStack initialization.

### Use Case

When deploying large asset bundles around of 3 GB (e.g., Lambda function packages, website files) to Service Catalog products, the internal Lambda function fails due to insufficient ephemeral storage.

For example:

- "I’m frustrated when large files (>500 MB) cause the internal Lambda function to fail with ENOSPC (no space left on device), and I cannot adjust the ephemeral storage or other limits."
- "Currently, there’s no way to address these issues without extensive custom workarounds that re-implement the CDK logic for asset management."
This feature would significantly improve flexibility for handling larger workloads in Service Catalog deployments.

### Proposed Solution

Add an optional property to the servicecatalog.ProductStackProps that allows customization of the internal Lambda function.

```
interface ProductStackProps {
...
lambdaProps?: lambda.FunctionProps; // Allow custom configuration
}
```

Developers can pass lambdaProps to override settings for the internal Lambda, such as:
```
new servicecatalog.ProductStack(this, 'TestProductStack', {
assetBucket: userDefinedBucket,
lambdaProps: {
memorySize: 1024,
ephemeralStorageSize: Size.gibibytes(3), // Customize storage
},
});
```

### Other Information

The error that I'm facing,
```
download failed: s3://cdk-hnb659fds-assets-1234567899-us-east-2/f38cbf44b36acd518668341f19c50896575a52313f6ea12dcad359f3fccd9978.zip to ../../tmp/tmp44wplx45/contents/f38cbf44b36acd518668341f19c50896575a52313f6ea12dcad359f3fccd9978.zip [Errno 28] No space left on device
[ERROR] 2024-12-02T12:01:14.436Z 5e3f2fc5-6474-4ebr-be6d-e63abb99561f Command '['/opt/awscli/aws', 's3', 'cp', 's3://cdk-hnb659fds-assets-1234567899-us-east-2/f38cbf44b36acd518668341f19c50896575a52313f6ea12dcad359f3fccd9978.zip', '/tmp/tmp44wplx45/contents']' returned non-zero exit status 1.
Traceback (most recent call last):
File "/var/task/index.py", line 130, in handler
s3_deploy(s3_source_zips, s3_dest, user_metadata, system_metadata, prune, exclude, include, source_markers, extract)
File "/var/task/index.py", line 195, in s3_deploy
aws_command("s3", "cp", s3_source_zip, contents_dir)
File "/var/task/index.py", line 259, in aws_command
subprocess.check_call([aws] + list(args))
File "/var/lang/lib/python3.9/subprocess.py", line 373, in check_call
raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['/opt/awscli/aws', 's3', 'cp', 's3://cdk-hnb659fds-assets-1234567899-us-east-2/f38cbf44b36acd518668341f19c50896575a52313f6ea12dcad359f3fccd9978.zip', '/tmp/tmp44wplx45/contents']' returned non-zero exit status 1.
```
Ref: https://docs.aws.amazon.com/cdk/api/v2/docs/aws-cdk-lib.aws_servicecatalog.ProductStackProps.html

### Acknowledgements

- [ ] I may be able to implement this feature request
- [ ] This feature might incur a breaking change

### CDK version used

2.170.0

### Environment details (OS name and version, etc.)

Ubuntu 22.04.5 LTS

Start at the aws_servicecatalog ProductStack and ProductStackProps entry points, then trace the internal Lambda used for asset copying. The work is done when an optional lambdaProps configuration can set ephemeralStorageSize for that Lambda and large Service Catalog asset deployments no longer fail from insufficient temporary storage.

Written by the indexing model from the issue text.