Foundry template deployments on a resource that already exists fail

### Is there an existing issue for this?

- [x] I have searched the existing issues

### Describe the bug

Upon running a solution with a Foundry resource, which already exists in Azure, I get:

Content:
{"error":{"code":"InvalidTemplateDeployment","message":"The template deployment 'foundry' is not valid according to the validation procedure. The following resource provider(s) - 'Microsoft.CognitiveServices/accounts (2025-09-01)' reported preflight validation errors. Tracking id is '99dec1a9-1a89-4d41-80f7-7bdddbe4018b'. See inner errors for details.","details":[{"code":"UpdatingCustomDomainNotAllowed","message":"Updating or disabling sub domain is not supported."}]}}

This happens on deploy too, but since I don't need to change the Foundry as it's already in use, this is tolerable. Run mode however doesn't work, as everything else is dependent on the Foundry.

I haven't changed anything on this Foundry at the management plane level, and definitely not anything to do with subdomains.

### Expected Behavior

_No response_

### Steps To Reproduce

- Create any C# AppHost with a Foundry resource and dependent NodeApp resources. Ensure that the Foundry it deploys already exists
- `aspire start`
- Error message is logged on foundry resource start

### Exceptions (if any)

_No response_

### Aspire doctor output

_No response_

### Anything else?

_No response_

Reproduce the failure with an AppHost containing an existing Foundry resource and dependent NodeApp resources by running `aspire start`. Trace the Foundry resource's startup and deployment path to determine why it attempts the unsupported subdomain update; done means an existing Foundry starts without the validation error and its dependent resources can run.

Written by the indexing model from the issue text.