CLI crashes with panic if Microcks API returns unexpected responses

Nessuno ha ancora preso questa issue.

While using the CLI, I noticed that pkg/connectors/microcks_client.go uses panic(err) in several places when reading HTTP response bodies or unmarshaling JSON.

If there is a temporary network glitch (e.g., the connection drops while reading the body) or if the Microcks server returns an invalid JSON response (like a 502 Bad Gateway HTML page from a proxy), the CLI will abruptly crash and print a Go stack trace instead of handling the error gracefully.

Additionally, methods like CreateTestResult use unsafe type assertions on the parsed JSON map (e.g., createTestResp["id"].(string)), which will also cause a panic if the id field is missing from the response payload.

Expected Behavior
The CLI should catch these errors, bubble them up through the standard error return values, and print a clean, user-friendly error message to the terminal before exiting with a non-zero status code.

Actual Behavior
The CLI completely crashes with a stack trace.

Suggested Fix
We should replace all instances of panic(err) in these client methods with proper context-wrapped errors (e.g., fmt.Errorf("failed to read response: %w", err)) since the functions already have an error return type defined. We should also add the comma-ok idiom to the type assertions to ensure we don't crash on missing fields.

Inizia in pkg/connectors/microcks_client.go esaminando i metodi del client che leggono i body delle response, fanno il parsing del JSON e usano la type assertion id. Sostituisci i percorsi che causano un crash con errori restituiti e contestualizzati, e gestisci i campi della response mancanti o non validi; il lavoro è completato quando gli errori imprevisti di rete, JSON o struttura della response arrivano alla CLI come errori puliti con un codice di uscita diverso da zero, invece di uno stack trace.

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