[Bug]: F.grid_sample当前不支持mode=“bicubic”，导致glm-4.1v-thinking网络跑不通

### Checklist - [x] 1. 我已经搜索过相关问题，但没有得到预期的帮助. (https://gitee.com/mindspore/mindspore/issues) - [x] 2. 最新版本中该错误尚未修复. - [x] 3. 请注意，如果您提交的Bug描述缺少相应的环境信息和最小可复现的demo，我们将很难复现和解决该问题，从而降低收到反馈的可能性，甚至该问题将被关闭. ### 🐞 问题详细描述 F.grid_sample当前不支持mode=“bicubic”，导致glm-4.1v-thinking网络跑不通 #!/bin/bash find . | grep -E "(__pycache__|\.pyc|\.pyo$)" | xargs rm -rf source /usr/local/Ascend/ascend-toolkit/set_env.sh # 该变量只用于规避megatron对其校验，对npu无效 export CUDA_DEVICE_MAX_CONNECTIONS=1 export ASCEND_SLOG_PRINT_TO_STDOUT=0 export ASCEND_GLOBAL_LOG_LEVEL=3 export TASK_QUEUE_ENABLE=2 export COMBINED_ENABLE=1 export CPU_AFFINITY_CONF=1 export HCCL_CONNECT_TIMEOUT=1200 export NPU_ASD_ENABLE=0 export ACLNN_CACHE_LIMIT=100000 export PYTORCH_NPU_ALLOC_CONF="expandable_segments:True" : ${MindSpeed_Core_MS_PATH:=./MindSpeed-Core-MS} echo ${MindSpeed_Core_MS_PATH} export PYTHONPATH=${MindSpeed_Core_MS_PATH}/Megatron-LM:${MindSpeed_Core_MS_PATH}/MindSpeed:${MindSpeed_Core_MS_PATH}/transformers_glm4/src:$PYTHONPATH echo $PYTHONPATH export HCCL_DETERMINISTIC=true # HCCL确定性 export ASCEND_LAUNCH_BLOCKING=1 # 硬件确定性 export NCCL_DETERMINISTIC=1 export CLOSE_MATMUL_K_SHIFT=1 #设置matmul行为 # export LD_PRELOAD="/root/miniconda3/envs/pt_mm_sl/lib/python3.10/site-packages/sklearn/utils/../../scikit_learn.libs/libgomp-d22c30c5.so.1.0.0":$LD_PRELOAD NPUS_PER_NODE=8 MASTER_ADDR=localhost MASTER_PORT=8467 NNODES=1 NODE_RANK=0 WORLD_SIZE=$(($NPUS_PER_NODE*$NNODES)) export LOCAL_WORLD_SIZE=8 MM_DATA="./examples/glm4.1v/data_9b.json" MM_MODEL="./examples/glm4.1v/model_9b.json" MM_TOOL="./mindspeed_mm/tools/tools.json" SAVE_PATH="./mm/ckpts/glm41v_8p" TP=1 PP=4 CP=1 MBS=1 GRAD_ACC_STEP=64 DP=$(($WORLD_SIZE/$TP/$PP/$CP)) GBS=$(($MBS*$GRAD_ACC_STEP*$DP)) DISTRIBUTED_ARGS=" --nproc_per_node $NPUS_PER_NODE \ --nnodes $NNODES \ --node_rank $NODE_RANK \ --master_addr $MASTER_ADDR \ --master_port $MASTER_PORT " GPT_ARGS=" --use-mcore-models \ --tensor-model-parallel-size ${TP} \ --pipeline-model-para…