Added TensorMNIST to speed up computation, minor fix in agem and mas

[AndreaCossu]

The main change of the PR is related to the default MNIST dataset, which is now called TorchMNIST. The dataset does not return a PIL image anymore but directly a Torch tensor.
This speeds up computation of strategies like A-GEM which access dataset items multiple times. The main bottleneck is due to the ToPIL transformation present in the default MNIST implementation (which requires a subsequent ToTensor transformation, thus increasing the total time).

[ContinualAI-bot]

Oh no! It seems there are some PEP8 errors! 😕
Don't worry, you can fix them! 💪
Here's a report about the errors and where you can find them:

avalanche/training/plugins/agem.py:8:81: E501 line too long (83 > 80 characters)
avalanche/training/plugins/agem.py:34:81: E501 line too long (83 > 80 characters)
avalanche/training/plugins/mas.py:9:81: E501 line too long (86 > 80 characters)
avalanche/training/plugins/mas.py:112:66: E261 at least two spaces before inline comment
avalanche/benchmarks/datasets/external_datasets/mnist.py:15:81: E501 line too long (92 > 80 characters)
avalanche/benchmarks/datasets/external_datasets/mnist.py:39:18: E128 continuation line under-indented for visual indent
avalanche/benchmarks/utils/data_loader.py:315:1: E303 too many blank lines (3)
1       E128 continuation line under-indented for visual indent
1       E261 at least two spaces before inline comment
1       E303 too many blank lines (3)
4       E501 line too long (83 > 80 characters)

[coveralls]

Pull Request Test Coverage Report for Build 3584722552

• 30 of 34 (88.24%) changed or added relevant lines in 4 files are covered.
• 4 unchanged lines in 3 files lost coverage.
• Overall coverage decreased (-0.02%) to 73.731%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
avalanche/training/plugins/mas.py	13	14	92.86%
avalanche/benchmarks/datasets/external_datasets/mnist.py	11	14	78.57%

Files with Coverage Reduction	New Missed Lines	%
avalanche/evaluation/metrics/images_samples.py	1	84.62%
avalanche/benchmarks/datasets/external_datasets/mnist.py	1	83.33%
avalanche/benchmarks/scenarios/classification_scenario.py	2	86.31%

Totals
Change from base Build 3575187291:	-0.02%
Covered Lines:	13652
Relevant Lines:	18516

---

💛 - Coveralls