Adding API key support

Summary:

Test Plan:
Try to config with no args
```
llama-stack-client configure
> Enter the host name of the Llama Stack distribution server: api.hostname.com
> Enter the port number of the Llama Stack distribution server: 80
> Enter the API key (leave empty if no key is needed): _API_KEY_
> Is Llama Stack distribution server using HTTPS? (y/n): n

Done! You can now use the Llama Stack Client CLI with endpoint https://api.hostname.com:80
```

Config with args
```
llama-stack-client configure --endpoint=https://api.hostname.com --api-key='_API_KEY_'

Done! You can now use the Llama Stack Client CLI with endpoint https://api.hostname.com
```

Try listing the models
```
llama-stack-client models list

┏━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━━┓
┃ identifier            ┃ provider_id ┃ provider_resource_id  ┃ metadata ┃ model_type ┃
┡━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━━┩
│ llama3.1-8b-instruct  │ meta-llama  │ llama3.1-8b-instruct  │ {}       │ llm        │
│ llama3.3-70b-instruct │ meta-llama  │ llama3.3-70b-instruct │ {}       │ llm        │
│ llama3.4-17b-instruct   │ meta-llama  │ llama3.4-17b-instruct   │ {}       │ llm        │
└───────────────────────┴─────────────┴───────────────────────┴──────────┴────────────┘
```

Author: vladimirivic

src/llama_stack_client/_client.py

@@ -103,6 +103,7 @@ def __init__(
         self,
         *,
         base_url: str | httpx.URL | None = None,
+        api_key: str | None = None,
         timeout: Union[float, Timeout, None, NotGiven] = NOT_GIVEN,
         max_retries: int = DEFAULT_MAX_RETRIES,
         default_headers: Mapping[str, str] | None = None,
@@ -132,6 +133,8 @@ def __init__(
         custom_headers["X-LlamaStack-Client-Version"] = __version__
         if provider_data is not None:
             custom_headers["X-LlamaStack-Provider-Data"] = json.dumps(provider_data)
+        if api_key is not None:
+            custom_headers["Authorization"] = f"Bearer {api_key}"
 
         super().__init__(
             version=__version__,
@@ -300,6 +303,7 @@ def __init__(
         self,
         *,
         base_url: str | httpx.URL | None = None,
+        api_key: str | None = None,
         timeout: Union[float, Timeout, None, NotGiven] = NOT_GIVEN,
         max_retries: int = DEFAULT_MAX_RETRIES,
         default_headers: Mapping[str, str] | None = None,
@@ -329,6 +333,8 @@ def __init__(
         custom_headers["X-LlamaStack-Client-Version"] = __version__
         if provider_data is not None:
             custom_headers["X-LlamaStack-Provider-Data"] = json.dumps(provider_data)
+        if api_key is not None:
+            custom_headers["Authorization"] = f"Bearer {api_key}"
 
         super().__init__(
             version=__version__,

src/llama_stack_client/lib/cli/configure.py

@@ -26,8 +26,10 @@ def get_config():
 @click.command()
 @click.option("--host", type=str, help="Llama Stack distribution host")
 @click.option("--port", type=str, help="Llama Stack distribution port number")
+@click.option("--use-https", type=bool, default=False, help="Llama Stack distribution uses HTTPS")
 @click.option("--endpoint", type=str, help="Llama Stack distribution endpoint")
-def configure(host: str | None, port: str | None, endpoint: str | None):
+@click.option("--api-key", type=str, help="Llama Stack distribution API key")
+def configure(host: str | None, port: str | None, use_https: bool | None, endpoint: str | None, api_key: str | None):
     """Configure Llama Stack Client CLI"""
     os.makedirs(LLAMA_STACK_CLIENT_CONFIG_DIR, exist_ok=True)
     config_path = get_config_file_path()
@@ -36,7 +38,10 @@ def configure(host: str | None, port: str | None, endpoint: str | None):
         final_endpoint = endpoint
     else:
         if host and port:
-            final_endpoint = f"http://{host}:{port}"
+            if use_https:
+                final_endpoint = f"https://{host}:{port}"
+            else:
+                final_endpoint = f"http://{host}:{port}"
         else:
             host = prompt(
                 "> Enter the host name of the Llama Stack distribution server: ",
@@ -52,14 +57,33 @@ def configure(host: str | None, port: str | None, endpoint: str | None):
                     error_message="Please enter a valid port number",
                 ),
             )
-            final_endpoint = f"http://{host}:{port}"
+            api_key = prompt(
+                "> Enter the API key (leave empty if no key is needed): ",
+            )
+
+            is_https = prompt(
+                "> Is Llama Stack distribution server using HTTPS? (y/n): ",
+                validator=Validator.from_callable(
+                    lambda x: x.lower() in ["y", "n", "yes", "no"],
+                    error_message="Please enter a valid response for HTTPS, yes or no",
+                ),
+            )
+            if is_https == "y":
+                final_endpoint = f"https://{host}:{port}"
+            else:
+                final_endpoint = f"http://{host}:{port}"
+
+    # Prepare config dict before writing it
+    config_dict = {
+        "endpoint": final_endpoint,
+    }
+    if api_key:
+        config_dict["api_key"] = api_key
 
     with open(config_path, "w") as f:
         f.write(
             yaml.dump(
-                {
-                    "endpoint": final_endpoint,
-                },
+                config_dict,
                 sort_keys=True,
             )
         )

src/llama_stack_client/lib/cli/llama_stack_client.py

@@ -55,15 +55,21 @@ def cli(ctx, endpoint: str, config: str | None):
             with open(config, "r") as f:
                 config_dict = yaml.safe_load(f)
                 endpoint = config_dict.get("endpoint", endpoint)
+                api_key = config_dict.get("api_key", None)
         except Exception as e:
             click.echo(f"Error loading config from {config}: {str(e)}", err=True)
             click.echo("Falling back to HTTP client with endpoint", err=True)
 
     if endpoint == "":
         endpoint = "http://localhost:8321"
 
+    print(f"[DEBUG] Using base url: {endpoint}")
+    if api_key:
+        print(f"[DEBUG] Using API key: {api_key}")
+
     client = LlamaStackClient(
         base_url=endpoint,
+        api_key=api_key,
         provider_data={
             "fireworks_api_key": os.environ.get("FIREWORKS_API_KEY", ""),
             "together_api_key": os.environ.get("TOGETHER_API_KEY", ""),