[Trainer] dictionary access to multiple named loggers

Nobody has claimed this yet.

## 🚀 Feature

When using multiple loggers, being able to give a dict of loggers to the `Trainer()` instead of a list.

### Motivation

I find the current list access when calling a single logger not explicit.

### Pitch

```python
trainer = Trainer(
logger={
'tensorboard': TensorBoardLogger(...),
'csv': CSVLogger(...)},
...
)
```
then in the loop:
```python
self.logger['tensorboard'].experiment.add_image(...)
```
instead of
```python
self.logger[0].experiment.add_image(...)
```

There could be an ambiguity when using an integer as key, so maybe enforce string-only keys.

cc @borda @awaelchli @edward-io @ananthsub @rohitgr7 @kamil-kaczmarek @Raalsky @Blaizzy @justusschock @kaushikb11

Start at the Trainer() logger argument and follow how the current list of loggers becomes self.logger. Compare the proposed dictionary access for named TensorBoardLogger and CSVLogger instances, including the unresolved integer-key ambiguity. Done means multiple named loggers can be passed and accessed by string key without breaking existing logger behavior.

Written by the indexing model from the issue text.