fix: multiple issues with getting_started notebook (#1795)

Fixes multiple issues 

1. llama stack build of dependencies was breaking with incompatible
numpy / pandas when importing datasets

Moved the notebook to start a local server instead of using library as a
client. This way the setup is cleaner since its all contained and by
using `uv run --with` we can test both the server setup process too in
CI and release time.

2. The change to [1] surfaced some other issues 
- running `llama stack run` was defaulting to conda env name 
- provider data was not being managed properly 
- Some notebook cells (telemetry for evals) were not updated with latest
changes

Fixed all the issues and update the notebook. 

### Test 

1. Manually run it all in local env 
2. `pytest -v -s --nbval-lax docs/getting_started.ipynb`

llama_stack/cli/stack/run.py

@@ -43,7 +43,7 @@ class StackRun(Subcommand):
         self.parser.add_argument(
             "--image-name",
             type=str,
-            default=os.environ.get("CONDA_DEFAULT_ENV"),
+            default=None,
             help="Name of the image to run. Defaults to the current conda environment",
         )
         self.parser.add_argument(

llama_stack/distribution/start_stack.sh

@@ -13,6 +13,7 @@ LLAMA_CHECKPOINT_DIR=${LLAMA_CHECKPOINT_DIR:-}
 LLAMA_STACK_DIR=${LLAMA_STACK_DIR:-}
 TEST_PYPI_VERSION=${TEST_PYPI_VERSION:-}
 PYPI_VERSION=${PYPI_VERSION:-}
+VIRTUAL_ENV=${VIRTUAL_ENV:-}
 
 set -euo pipefail
 
@@ -69,22 +70,25 @@ while [[ $# -gt 0 ]]; do
     ;;
   esac
 done
-
 PYTHON_BINARY="python"
 case "$env_type" in
   "venv")
-    # Activate virtual environment
+    if [ -n "$VIRTUAL_ENV" && "$VIRTUAL_ENV" == "$env_path_or_name" ]; then
-    if [ ! -d "$env_path_or_name" ]; then
+        echo -e "${GREEN}Virtual environment already activated${NC}" >&2
-        echo -e "${RED}Error: Virtual environment not found at $env_path_or_name${NC}" >&2
+    else
-        exit 1
+        # Activate virtual environment
-    fi
+        if [ ! -d "$env_path_or_name" ]; then
+            echo -e "${RED}Error: Virtual environment not found at $env_path_or_name${NC}" >&2
+            exit 1
+        fi
 
-    if [ ! -f "$env_path_or_name/bin/activate" ]; then
+        if [ ! -f "$env_path_or_name/bin/activate" ]; then
-        echo -e "${RED}Error: Virtual environment activate binary not found at $env_path_or_name/bin/activate" >&2
+            echo -e "${RED}Error: Virtual environment activate binary not found at $env_path_or_name/bin/activate" >&2
-        exit 1
+            exit 1
-    fi
+        fi
 
-    source "$env_path_or_name/bin/activate"
+        source "$env_path_or_name/bin/activate"
+    fi
     ;;
   "conda")
     if ! is_command_available conda; then

llama_stack/distribution/utils/context.py

@@ -18,15 +18,19 @@ def preserve_contexts_async_generator(
     This is needed because we start a new asyncio event loop for each streaming request,
     and we need to preserve the context across the event loop boundary.
     """
+    # Capture initial context values
+    initial_context_values = {context_var.name: context_var.get() for context_var in context_vars}
 
     async def wrapper() -> AsyncGenerator[T, None]:
         while True:
             try:
-                item = await gen.__anext__()
+                # Restore context values before any await
-                context_values = {context_var.name: context_var.get() for context_var in context_vars}
-                yield item
                 for context_var in context_vars:
-                    _ = context_var.set(context_values[context_var.name])
+                    context_var.set(initial_context_values[context_var.name])
+
+                item = await gen.__anext__()
+                yield item
+
             except StopAsyncIteration:
                 break