Add support for ClearML logging

Nobody has claimed this yet.

## 🚀 Feature

Given the popularity of [ClearML](https://clear.ml) it would be great to have integration with it for experiment tracking realized trough a logger.

### Motivation

At the time being many MLOps platforms are appearing and gaining popularity. On the other side, the architecture of PyTorch Lightning, primarily through the abstractions enabling integrations with new platforms is easier mainly following the Open-Closed Principle.

### Pitch

If this feature request proceeds forward there would be seamless integration with ClearML following the examples for WANDB and MLFlow.

### Alternatives

Alternative would be coupling the training code for calls to the API provided by the [`clearml`](https://pypi.org/project/clearml/) package for achieving experiment tracking.

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

Start by reviewing the existing WANDB and MLFlow logger integrations, then read the ClearML package documentation to understand the experiment-tracking API. Done means providing a ClearML logger integration comparable to those examples, with the requested experiment-tracking support working through PyTorch Lightning.

Written by the indexing model from the issue text.