feat: drop python 3.10 support (#2469)

# What does this PR do?

dropped python3.10, updated pyproject and dependencies, and also removed
some blocks of code with special handling for enum.StrEnum

Closes #2458

Signed-off-by: Charlie Doern <cdoern@redhat.com>

llama_stack/distribution/access_control/datatypes.py

@@ -5,9 +5,9 @@
 # the root directory of this source tree.
 
 from enum import Enum
+from typing import Self
 
 from pydantic import BaseModel, model_validator
-from typing_extensions import Self
 
 from .conditions import parse_conditions
 

llama_stack/distribution/build.py

@@ -101,7 +101,7 @@ def build_image(
     template_or_config: str,
     run_config: str | None = None,
 ):
-    container_base = build_config.distribution_spec.container_image or "python:3.10-slim"
+    container_base = build_config.distribution_spec.container_image or "python:3.11-slim"
 
     normal_deps, special_deps = get_provider_dependencies(build_config)
     normal_deps += SERVER_DEPENDENCIES

llama_stack/distribution/build_conda_env.sh

@@ -49,7 +49,7 @@ ensure_conda_env_python310() {
   local env_name="$1"
   local pip_dependencies="$2"
   local special_pip_deps="$3"
-  local python_version="3.10"
+  local python_version="3.11"
 
   # Check if conda command is available
   if ! is_command_available conda; then

llama_stack/distribution/providers.py

@@ -99,7 +99,7 @@ class ProviderImpl(Providers):
             try:
                 health = await asyncio.wait_for(impl.health(), timeout=timeout)
                 return api_name, health
-            except (asyncio.TimeoutError, TimeoutError):
+            except TimeoutError:
                 return (
                     api_name,
                     HealthResponse(

llama_stack/distribution/routers/inference.py

@@ -615,7 +615,7 @@ class InferenceRouter(Inference):
                     continue
                 health = await asyncio.wait_for(impl.health(), timeout=timeout)
                 health_statuses[provider_id] = health
-            except (asyncio.TimeoutError, TimeoutError):
+            except TimeoutError:
                 health_statuses[provider_id] = HealthResponse(
                     status=HealthStatus.ERROR,
                     message=f"Health check timed out after {timeout} seconds",

llama_stack/distribution/server/auth_providers.py

@@ -9,12 +9,12 @@ import time
 from abc import ABC, abstractmethod
 from asyncio import Lock
 from pathlib import Path
+from typing import Self
 from urllib.parse import parse_qs
 
 import httpx
 from jose import jwt
 from pydantic import BaseModel, Field, field_validator, model_validator
-from typing_extensions import Self
 
 from llama_stack.distribution.datatypes import AuthenticationConfig, AuthProviderType, User
 from llama_stack.log import get_logger

llama_stack/distribution/server/quota.py

@@ -6,7 +6,7 @@
 
 import json
 import time
-from datetime import datetime, timedelta, timezone
+from datetime import UTC, datetime, timedelta
 
 from starlette.types import ASGIApp, Receive, Scope, Send
 
@@ -79,7 +79,7 @@ class QuotaMiddleware:
 
                 if int(prev) == 0:
                     # Set with expiration datetime when it is the first request in the window.
-                    expiration = datetime.now(timezone.utc) + timedelta(seconds=self.window_seconds)
+                    expiration = datetime.now(UTC) + timedelta(seconds=self.window_seconds)
                     await kv.set(key, str(count), expiration=expiration)
                 else:
                     await kv.set(key, str(count))

llama_stack/distribution/server/server.py

@@ -145,7 +145,7 @@ async def shutdown(app):
                 await asyncio.wait_for(impl.shutdown(), timeout=5)
             else:
                 logger.warning("No shutdown method for %s", impl_name)
-        except (asyncio.TimeoutError, TimeoutError):
+        except TimeoutError:
             logger.exception("Shutdown timeout for %s ", impl_name, exc_info=True)
         except (Exception, asyncio.CancelledError) as e:
             logger.exception("Failed to shutdown %s: %s", impl_name, {e})