Bug: namespace deletion metrics always recorded with empty project label due to double GetProjectName call

### 📜 Description

`NamespaceService.DeleteNamespace` at line 192 calls `util.GetProjectName(util.GetProjectName(namespace))`. The inner call correctly extracts the project name (e.g., `"cisco"` from `"kubeslice-cisco"`). The outer call receives `"cisco"` which lacks the `"kubeslice-"` prefix and returns `""`. All Prometheus counter metrics for namespace deletion events carry `project=""`. Every other method in `namespace_service.go` calls `GetProjectName` exactly once; this is the only double-wrapped call.

### 👟 Reproduction steps

1. Read `service/namespace_service.go:192` — `n.mf.WithProject(util.GetProjectName(util.GetProjectName(namespace)))`. Compare with every other `WithProject` call in the same file; all use a single `GetProjectName`.
2. At runtime:
- Delete any Project (which triggers `DeleteNamespace`).
- Query Prometheus: `kubeslice_controller_events_total{event="EventNamespaceDeleted"}`.
3. The `project` label is `""` instead of the actual project name.

### 👍 Expected behavior

The `project` label on namespace deletion metrics should equal the actual project name extracted from the namespace string.

### 👎 Actual Behavior

`project=""` on all namespace deletion metrics because the double `GetProjectName` call strips the extracted name back to an empty string.

### 🐚 Relevant log output

```shell

```

### Version

_No response_

### 🖥️ What operating system are you seeing the problem on?

_No response_

### ✅ Proposed Solution

Remove the outer `GetProjectName` wrapper on line 192 so the call reads `util.GetProjectName(namespace)`, matching the pattern used in every other method in the file.

### 👀 Have you spent some time to check if this issue has been raised before?

- [x] I checked and didn't find any similar issue

### Code of Conduct

- [x] I agree to follow this project's Code of Conduct