Implement simple VM manager with CRUD operations

You are asked to implement a simple **in-memory virtual machine (VM) manager** that can: - List all VMs - Add (create) a new VM - Modify (update) an existing VM - Delete an existing VM Each VM has a set of attributes. For this exercise, you can assume at least the following attributes: - `id` (string or integer, **must be unique**) - `name` (string) - `cpu_cores` (integer) - `memory_gb` (integer) - `status` (string, e.g., `"running"`, `"stopped"`) ### Requirements 1. **Data model & storage** - Choose an appropriate in-memory data structure to store the VMs. - Ensure `id` is treated as the unique identifier for each VM. 2. **API / Interface** Design and implement a small API (in the language of your choice) with at least the following operations: - `create_vm(vm)` - Input: VM attributes (including `id`). - Behavior: Adds a new VM to the manager. - `list_vms()` - Input: none. - Output: a collection/list of all VMs currently stored. - `update_vm(id, updated_fields)` - Input: the `id` of the VM to update, and a set/dictionary of fields to update. - Behavior: Modifies the attributes of the existing VM with the given `id`. - `delete_vm(id)` - Input: the `id` of the VM to delete. - Behavior: Removes the VM with the given `id` from the manager. 3. **Error handling** - If `create_vm` is called with an `id` that already exists, you must detect this and handle it as an error. - If `update_vm` or `delete_vm` is called with an `id` that does **not** exist, you must detect this and handle it as an error. - Define clearly how errors are surfaced (e.g., exceptions, error codes, or special return values) and keep it consistent across methods. 4. **Complexity considerations** - Aim for efficient average-time operations for lookup, update, and delete by `id`. Explain your design choices (data structures and API shape) in brief comments or docstrings within your code.