[SYCL] atomic_memory_order_capabilities query for device and context (intel#8517)

This patch implements the `atomic_memory_order_capabilities` query in
the OpenCL and Level Zero backends/plugins for `device` and `context`

Specifically: 
- OpenCL <2.0 returns the minimum required capability set (`relaxed`)
defined in [Section 4.2 of the OpenCL 3.0
specification](https://registry.khronos.org/OpenCL/specs/3.0-unified/html/OpenCL_API.html#CL_DEVICE_ATOMIC_MEMORY_CAPABILITIES).
- OpenCL <3.0 and Level Zero backends return all memory order
capabilities.
- OpenCL >=3.0 queries the actual device to get the supported memory
order capabilities.

E2E test have also been updated to reflect these changes:
intel/llvm-test-suite#1627

Author: Andrew Lamzed-Short

ur_level_zero.cpp

@@ -1186,6 +1186,17 @@ UR_APIEXPORT ur_result_t UR_APICALL urDeviceGetInfo(
     return ReturnValue(result);
   }
 
+  case UR_DEVICE_INFO_ATOMIC_MEMORY_ORDER_CAPABILITIES: {
+    ur_memory_order_capability_flags_t capabilities =
+        UR_MEMORY_ORDER_CAPABILITY_FLAG_RELAXED |
+        UR_MEMORY_ORDER_CAPABILITY_FLAG_ACQUIRE |
+        UR_MEMORY_ORDER_CAPABILITY_FLAG_RELEASE |
+        UR_MEMORY_ORDER_CAPABILITY_FLAG_ACQ_REL |
+        UR_MEMORY_ORDER_CAPABILITY_FLAG_SEQ_CST;
+    return ReturnValue(capabilities);
+  }
+
+  // TODO: Implement.
   default:
     zePrint("Unsupported ParamName in piGetDeviceInfo\n");
     zePrint("ParamName=%d(0x%x)\n", ParamName, ParamName);
@@ -1716,7 +1727,7 @@ UR_APIEXPORT ur_result_t UR_APICALL urDevicePartition(
   // Currently supported partitioning (by affinity domain/numa) would always
   // partition to all sub-devices.
   //
-  if (NumDevices !=0)
+  if (NumDevices != 0)
     PI_ASSERT(NumDevices == EffectiveNumDevices, UR_RESULT_ERROR_INVALID_VALUE);
 
   for (uint32_t I = 0; I < NumDevices; I++) {