Different DDP ranks have different bn_stats after `PreciseBN` hook as precise_bn in fvcore does not synchronize batch_size

## Instructions To Reproduce the 🐛 Bug:
1. Full runnable code or full changes you made:
https://github.com/facebookresearch/moco/tree/main/detection
2. What exact command you run:

```python
python detection/train_net.py \
--config-file detection/configs/pascal_voc_R_50_C4_24k.yaml \
--num-gpus 8 \
OUTPUT_DIR "temp/train" \
SEED 0 \
SOLVER.MAX_ITER 1
```

3. __Full logs__ or other relevant observations:

In `update_bn_stats`, different ranks have different batch_size as it's not synchronized, resulting in different bn_stats on different ranks after `PreciseBN` hook.

4. please simplify the steps as much as possible so they do not require additional resources to
run, such as a private dataset.

no private dataset

## Expected behavior:

Same bn_stats after `PreciseBN` hook.

## Environment:

Provide your environment information using the following command:
```
wget -nc -q https://github.com/facebookresearch/detectron2/raw/main/detectron2/utils/collect_env.py && python collect_env.py
```

```
------------------------------- ---------------------------------------------------------------------------------------------
sys.platform linux
Python 3.8.0 | packaged by conda-forge | (default, Nov 22 2019, 19:11:38) [GCC 7.3.0]
numpy 1.24.4
detectron2 0.6 @/mdata/guzy0324/anaconda3/envs/DICR_new/lib/python3.8/site-packages/detectron2
Compiler GCC 11.4
CUDA compiler CUDA 12.4
detectron2 arch flags 8.9
DETECTRON2_ENV_MODULE
PyTorch 2.4.1+cu124 @/mdata/guzy0324/anaconda3/envs/DICR_new/lib/python3.8/site-packages/torch
PyTorch debug build False
torch._C._GLIBCXX_USE_CXX11_ABI False
GPU available Yes
GPU 0,1,2,3,4,5,6,7 NVIDIA GeForce RTX 4090 D (arch=8.9)
Driver version 550.54.14
CUDA_HOME /usr/local/cuda
Pillow 9.4.0
torchvision 0.19.1+cu124 @/mdata/guzy0324/anaconda3/envs/DICR_new/lib/python3.8/site-packages/torchvision
torchvision arch flags 5.0, 6.0, 7.0, 7.5, 8.0, 8.6, 9.0
fvcore 0.1.5.post20221221
iopath 0.1.9
cv2 4.9.0
------------------------------- ---------------------------------------------------------------------------------------------
PyTorch built with:
- GCC 9.3
- C++ Version: 201703
- Intel(R) oneAPI Math Kernel Library Version 2022.2-Product Build 20220804 for Intel(R) 64 architecture applications
- Intel(R) MKL-DNN v3.4.2 (Git Hash 1137e04ec0b5251ca2b4400a4fd3c667ce843d67)
- OpenMP 201511 (a.k.a. OpenMP 4.5)
- LAPACK is enabled (usually provided by MKL)
- NNPACK is enabled
- CPU capability usage: AVX512
- CUDA Runtime 12.4
- NVCC architecture flags: -gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_90,code=sm_90
- CuDNN 90.1
- Magma 2.6.1
- Build settings: BLAS_INFO=mkl, BUILD_TYPE=Release, CUDA_VERSION=12.4, CUDNN_VERSION=9.1.0, CXX_COMPILER=/opt/rh/devtoolset-9/root/usr/bin/c++, CXX_FLAGS= -D_GLIBCXX_USE_CXX11_ABI=0 -fabi-version=11 -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -DNDEBUG -DUSE_KINETO -DLIBKINETO_NOROCTRACER -DUSE_FBGEMM -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -O2 -fPIC -Wall -Wextra -Werror=return-type -Werror=non-virtual-dtor -Werror=bool-operation -Wnarrowing -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-stringop-overflow -Wsuggest-override -Wno-psabi -Wno-error=pedantic -Wno-error=old-style-cast -Wno-missing-braces -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, LAPACK_INFO=mkl, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, TORCH_VERSION=2.4.1, USE_CUDA=ON, USE_CUDNN=ON, USE_CUSPARSELT=1, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_GLOO=ON, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=1, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, USE_ROCM_KERNEL_ASSERT=OFF,

Testing NCCL connectivity ... this should not hang.
NCCL succeeded.
```

If your issue looks like an installation issue / environment issue,
please first try to solve it yourself with the instructions in
https://detectron2.readthedocs.io/tutorials/install.html#common-installation-issues

Start with detection/train_net.py and the PreciseBN hook, then inspect fvcore's update_bn_stats path used by the reproducible eight-GPU command. Verify how batch_size is handled across DDP ranks; done means all ranks have identical bn_stats after the hook.

Written by the indexing model from the issue text.