test: code exec on mac (#1549)

Summary:
1. adds option to not use bwrap for code execution
2. disable bwrap when running tests on macs

Test Plan:
```
LLAMA_STACK_CONFIG=fireworks pytest -s -v tests/integration/agents/test_agents.py --safety-shield meta-llama/Llama-Guard-3-8B --text-model meta-llama/Llama-3.1-8B-Instruct
```

Verify code_interpreter result in logs

INFO 2025-03-11 08:10:39,858
llama_stack.providers.inline.agents.meta_reference.agent_instance:1032
agents: tool
call code_interpreter completed with result:
content='completed\n\n541\n' error_message=None error_code=None
         metadata=None

llama_stack/providers/inline/tool_runtime/code_interpreter/code_execution.py

@@ -76,6 +76,7 @@ class CodeExecutionRequest:
     only_last_cell_fail: bool = True
     seed: int = 0
     strip_fpaths_in_stderr: bool = True
+    use_bwrap: bool = True
 
 
 class CodeExecutor:
@@ -103,8 +104,6 @@ _set_seeds()\
 
         script = "\n\n".join([seeds_prefix] + [CODE_ENV_PREFIX] + scripts)
         with tempfile.TemporaryDirectory() as dpath:
-            bwrap_prefix = "bwrap " + generate_bwrap_command(bind_dirs=[dpath])
-            cmd = [*bwrap_prefix.split(), sys.executable, "-c", script]
             code_fpath = os.path.join(dpath, "code.py")
             with open(code_fpath, "w") as f:
                 f.write(script)
@@ -118,6 +117,13 @@ _set_seeds()\
                     MPLBACKEND="module://matplotlib_custom_backend",
                     PYTHONPATH=f"{DIRNAME}:{python_path}",
                 )
+
+                if req.use_bwrap:
+                    bwrap_prefix = "bwrap " + generate_bwrap_command(bind_dirs=[dpath])
+                    cmd = [*bwrap_prefix.split(), sys.executable, "-c", script]
+                else:
+                    cmd = [sys.executable, "-c", script]
+
                 stdout, stderr, returncode = do_subprocess(
                     cmd=cmd,
                     env=env,

llama_stack/providers/inline/tool_runtime/code_interpreter/code_interpreter.py

@@ -6,6 +6,7 @@
 
 
 import logging
+import os
 import tempfile
 from typing import Any, Dict, List, Optional
 
@@ -61,7 +62,9 @@ class CodeInterpreterToolRuntimeImpl(ToolsProtocolPrivate, ToolRuntime):
 
     async def invoke_tool(self, tool_name: str, kwargs: Dict[str, Any]) -> ToolInvocationResult:
         script = kwargs["code"]
-        req = CodeExecutionRequest(scripts=[script])
+        # Use environment variable to control bwrap usage
+        force_disable_bwrap = os.environ.get("DISABLE_CODE_SANDBOX", "").lower() in ("1", "true", "yes")
+        req = CodeExecutionRequest(scripts=[script], use_bwrap=not force_disable_bwrap)
         res = self.code_executor.execute(req)
         pieces = [res["process_status"]]
         for out_type in ["stdout", "stderr"]:

tests/integration/agents/test_agents.py

@@ -187,7 +187,7 @@ def test_builtin_tool_web_search(llama_stack_client_with_mocked_inference, agent
         messages=[
             {
                 "role": "user",
-                "content": "Search the web and tell me who the current CEO of Meta is.",
+                "content": "Search the web and tell me who the founder of Meta is.",
             }
         ],
         session_id=session_id,

tests/integration/conftest.py

@@ -6,12 +6,17 @@
 import inspect
 import itertools
 import os
+import platform
 import textwrap
 
 from dotenv import load_dotenv
 
+from llama_stack.log import get_logger
+
 from .report import Report
 
+logger = get_logger(__name__, category="tests")
+
 
 def pytest_configure(config):
     config.option.tbstyle = "short"
@@ -24,6 +29,10 @@ def pytest_configure(config):
         key, value = env_var.split("=", 1)
         os.environ[key] = value
 
+    if platform.system() == "Darwin":  # Darwin is the system name for macOS
+        os.environ["DISABLE_CODE_SANDBOX"] = "1"
+        logger.info("Setting DISABLE_CODE_SANDBOX=1 for macOS")
+
     if config.getoption("--report"):
         config.pluginmanager.register(Report(config))