[TLE] Add device intra node barrier

[Galaxy1458]

Summary

This PR adds support for GPU-to-GPU (device-to-device) barriers within a node in the TLE backend, enabling synchronization across multiple GPUs through the FlagCX runtime.

Changes

Python frontend

• Extend check_and_handle_device_intra_barrier() to recognize device barriers.
• Lower distributed_barrier(..., space="device", comm_ptr=...) into the tle.device_intra_barrier operation.

MLIR

Introduce a new operation:

• tle.device_intra_barrier for FlagCxOp

The operation represents an intra-node GPU-to-GPU synchronization barrier with configurable:

• barrier type (arrive, wait, sync)
• cooperative group scope
• barrier index
• memory ordering
• multimem mode

LLVM Lowering

Lower tle.device_intra_barrier to the corresponding FlagCX runtime APIs:

• flagcxIntraBarrierArriveS
• flagcxIntraBarrierWaitS
• flagcxIntraBarrierSyncS

The runtime function is selected based on the requested barrier type.

Motivation

This PR introduces a dedicated IR representation for intra-node GPU-to-GPU synchronization, instead of representing it as a generic distributed barrier. It also establishes the infrastructure for future multi-GPU synchronization primitives built on the FlagCX runtime.

Example:

 tle.distributed_barrier(
              comm_ptr=dev_comm_dptr, space="device", 
              group_kind="block", order="acqrel",barrier_kind="sync"
)