WandbLogger does not log checkpoints from multiple ModelCheckpoint callbacks, but only from last one

Nobody has claimed this yet.

## 🐛 Bug

As the title explains, WandbLogger does not log checkpoints from multiple ModelCheckpoint callbacks, but only from last one. When multiple ModelCheckpoint callbacks are used, only the last one provided to the trainer in `callbacks=[..., LastModelCheckpoint]` is logged in Wandb, even though all the checkpoints are created as expected locally. Is this intended to be like this or is it an issue?

### To Reproduce
```
import os
import torch
from torch.utils.data import DataLoader, Dataset
from pytorch_lightning import LightningModule, Trainer
from pytorch_lightning.callbacks import ModelCheckpoint
from pytorch_lightning.loggers import WandbLogger

class RandomDataset(Dataset):
def __init__(self, size, length):
self.len = length
self.data = torch.randn(length, size)

def __getitem__(self, index):
return self.data[index]

def __len__(self):
return self.len

class BoringModel(LightningModule):
def __init__(self):
super().__init__()
self.layer = torch.nn.Linear(32, 2)

def forward(self, x):
return self.layer(x)

def training_step(self, batch, batch_idx):
loss = self(batch).sum()
self.log("train_loss", loss)
return {"loss": loss}

def validation_step(self, batch, batch_idx):
loss = self(batch).sum()
self.log("val_loss", loss)

def test_step(self, batch, batch_idx):
loss = self(batch).sum()
self.log("test_loss", loss)

def configure_optimizers(self):
return torch.optim.SGD(self.layer.parameters(), lr=0.1)

def run():
train_data = DataLoader(RandomDataset(32, 64), batch_size=2)
val_data = DataLoader(RandomDataset(32, 64), batch_size=2)
test_data = DataLoader(RandomDataset(32, 64), batch_size=2)

model = BoringModel()

max_steps = 500
min_epochs = 10
callback_last_step = ModelCheckpoint(
verbose=False,
save_on_train_epoch_end=False,
dirpath="./checkpoints",
filename="{epoch}_{step}_{val_loss:.2f}_last_step",
every_n_train_steps=max_steps,
)
callback_epoch10 = ModelCheckpoint(
verbose=False,
save_on_train_epoch_end=False,
dirpath="./checkpoints",
filename="{epoch}_{step}_{val_loss:.2f}_epoch10",
every_n_epochs=min_epochs,
)
callback_topk = ModelCheckpoint(
verbose=False,
save_on_train_epoch_end=False,
dirpath="./checkpoints",
filename="{epoch}_{step}_{val_loss:.2f}_topk",
every_n_epochs=1,
monitor="val_loss",
mode="min",
save_top_k=3,
)
callbacks = [callback_topk, callback_last_step, callback_epoch10]
logger = WandbLogger(
project="MNIST", entity="aleqrc", name="test", log_model=True, mode="online"
)
trainer = Trainer(
min_epochs=min_epochs,
max_steps=max_steps,
callbacks=callbacks,
logger=logger,
limit_train_batches=50,
limit_val_batches=50,
enable_model_summary=False,
num_sanity_val_steps=0,
)

trainer.fit(model, train_dataloaders=train_data, val_dataloaders=val_data)

if __name__ == "__main__":
run()
```

### Expected behavior
The code should end successfully, create 5 checkpoints (3 ending with "topk", 1 ending with "epoch10" and 1 ending with "last_step"), and log all of them in Wandb, however only checkpoints created by the last ModelCheckpoint provided to the trainer are logged.

### Environment
- PyTorch Lightning Version: 1.5.4
- PyTorch Version: 1.10.0
- Python version: 3.8.10
- OS: Windows 11
- CUDA/cuDNN version: 11.3.1

### Additional context
I have also encountered this error on a Linux machine (Ubuntu), even though I did not test exactly this code snippet, but I have the same issue there as well.

cc @awaelchli @morganmcg1 @AyushExel @borisdayma @scottire