[TRTLLM-6445] feat: Enable AllReduce associated fusion patterns in Llama3/4.

* Added support for controlling fusion optimizations via environment variables.
* Applied AR+Residual + RMS_NORM + Quant fp4/fp8 fusion. This is also compatible with the speculative decoding capturing in these models.
* Some improvements for the two-shot allreduce kernel.
* Disable fusion for small models with a hidden size no greater than 4096 to avoid accuracy drop issues.

Signed-off-by: Yukun He <[email protected]>

Author: hyukn

cpp/tensorrt_llm/kernels/communicationKernels/allReduceFusionKernels.cu

@@ -520,7 +520,7 @@ __global__ void __launch_bounds__(1024) allreduce_fusion_kernel_oneshot_lamport(
 }
 
 template <AllReduceFusionPattern Pattern, typename DType, int NRanks, bool Fp32Acc>
-__global__ void allreduce_fusion_kernel_twoshot_sync(
+__global__ void __launch_bounds__(1024) allreduce_fusion_kernel_twoshot_sync(
     AllReduceFusionParams params, std::array<int, NRanks> begin_tokens, std::array<int, NRanks> token_num_per_ranks)
 {
     IndexHelper<DType> index_helper(params);