[NPU]Improvement performence for grpo_loss

[UserChen666]

Summary

This PR implements Triton kernel-level performance optimizations for the grpo_loss operator.

Details

1. Block Size (BLOCK_N) & Memory Coefficient Tuning
   Unified default tiling by increasing BLOCK_N from 2048 to 4096 across all kernels, improving per-batch computation granularity and reducing loop counts and kernel launch overhead.
   Adjusted memory multipliers downward to fit the NPU’s 192KB Unified Buffer (UB) capacity and enhance memory utilization:
   Softmax: 6.0 → 3.0
   Forward: 10.0 → 4.0
   Backward: 12.0 → 8.0
2. Computation Instruction Optimization (Reduce Divisions, Improve Instruction Efficiency)
   Precompute inv_temp = 1.0 / TEMPERATURE to replace multiple in-loop divisions with single multiplications, reducing floating-point latency.
   Simplified backward gradient expression: dlogp = dlogp * dloss * inv_temp instead of the original chained division, lowering the number of floating-point operations.
3. Loop & Compilation Optimizations
   Changed inner kernel loops from range to tl.static_range to provide explicit loop-unrolling hints to the compiler, optimizing instruction scheduling and pipeline efficiency.
   Explicit index type casting: INPUT_IDS cast to int32 to avoid implicit type conversion overhead on NPU.
4. Masking & Memory Access Optimization
   Unified use of the cols_mask variable to reuse memory access masks, reducing redundant calculations and improving memory access throughput.
   Simplified gradient calculation logic: (cols_idx - probs) * dlogp instead of tl.where branching, minimizing branch judgment overhead.

Testing Done

• Hardware Type: Atlas A800I
• run make test to ensure correctness
• run make checkstyle to ensure code style
• run make test-convergence to ensure convergence

[UserChen666]

@Tcc0403 please review, thank you.

[UserChen666]

@Tcc0403

[Tcc0403]

LGTM