Have evaluator model config work with project endpoint

**Is your feature request related to a problem? Please describe.**
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]

Customers usually run into the confusion of using project endpoint in evaluator model config. We should consider making the model config accept a project endpont too assuming the model deployment is in the same resource.

```
To be clear, the azure_endpoint in AzureOpenAIModelConfiguration(
# can only use the azure openai endpoint. If using foundry endpoint, there will be an error
azure_endpoint=os.environ["AZURE_OPENAI_ENDPOINT"],
azure_deployment=os.environ["AZURE_OPENAI_DEPLOYMENT"],
)

must be Azure OpenAI endpoint, not Azure AI Foundry project endpoint (which is reserved for the parameter called azure_ai_project.
```

**Describe the solution you'd like**
A clear and concise description of what you want to happen.

**Describe alternatives you've considered**
A clear and concise description of any alternative solutions or features you've considered.

**Additional context**
Add any other context or screenshots about the feature request here.

Inizia individuando AzureOpenAIModelConfiguration e il percorso di configurazione del modello valutatore che gestisce azure_endpoint e azure_ai_project. Riproduci l’errore documentato dell’endpoint di progetto Foundry, quindi verifica che un endpoint di progetto venga accettato quando la distribuzione si trova nella stessa risorsa, senza compromettere l’uso degli endpoint Azure OpenAI.

Scritto dal modello di indicizzazione a partire dal testo della issue.