transforms/loading: support reading numpy files (.npy, .npz)

[jonGuti13]

This PR is related to open-mmlab/mmsegmentation#3365.

Motivation

Add support for reading numpy images which will enable working with hyperspectral/multispectral databases.

Modification

I have modified mmcv/transforms/loading.py so as to correctly open Numpy files. Documentation file openmm/mmcv/docs/en/understand_mmcv/data_process.md has also been updated. I lack the capacity to correct
openmm/mmcv/docs/zh_cn/understand_mmcv/data_process.md.

BC-breaking (Optional)

No.

Use cases (Optional)

This PR is related to open-mmlab/mmsegmentation#3365 which aims to include HSI Drive 2.0 dataset, a hyperspectral database, to the repository.

Checklist

Before PR:

• I have read and followed the workflow indicated in the CONTRIBUTING.md to create this PR.
• Pre-commit or linting tools indicated in CONTRIBUTING.md are used to fix the potential lint issues.
• Bug fixes are covered by unit tests, the case that causes the bug should be added in the unit tests. (There is no bug so it does not apply)
• New functionalities are covered by complete unit tests. If not, please add more unit test to ensure the correctness (I think it applies neither)
• The documentation has been modified accordingly, including docstring or example tutorials. (Only partially as it is described in Modification).

After PR:

• If the modification has potential influence on downstream or other related projects, this PR should be tested with some of those projects, like MMDet or MMCls.
• CLA has been signed and all committers have signed the CLA in this PR.

[CLAassistant]

All committers have signed the CLA.

[kbasterr]

This would be great!