Use pip instead of conda to speed up Actions

[leouieda]

Switch to using the built-in Python and pip to setup Actions workflows
instead of conda/miniconda. This speeds up the workflows by a large
amount and is worth the trade-off since we don't have any dependencies
that require conda.

Reminders:

• Run make format and make check to make sure the code follows the style guide.
• Add tests for new features or tests that would have caught the bug that you're fixing.
• Add new public functions/methods/classes to doc/api/index.rst and the base __init__.py file for the package.
• Write detailed docstrings for all functions/classes/methods. It often helps to design better code if you write the docstrings first.
• If adding new functionality, add an example to the docstring, gallery, and/or tutorials.
• Add your full name, affiliation, and ORCID (optional) to the AUTHORS.md file (if you haven't already) in case you'd like to be listed as an author on the Zenodo archive of the next release.

[leouieda]

Tests now take around 1m to complete instead of 5-6minutes for Windows and Mac with conda.

[hugovk]

It's possible to cache pip stuff too, something like this:

      - name: Get pip cache dir
        id: pip-cache
        run: |
          echo "::set-output name=dir::$(pip cache dir)"

      - name: Cache
        uses: actions/cache@v2
        with:
          path: ${{ steps.pip-cache.outputs.dir }}
          key:
            pip-${{ runner.os }}-${{ matrix.python }}-${{ hashFiles('requirements*.txt') }}
          restore-keys: |
            pip-${{ runner.os }}-${{ matrix.python }}-

Might make things a bit quicker too, and help ease the strain on PyPI just a little bit.

I also saw in some logs:

Using legacy 'setup.py install' for pytest-localftpserver, since package 'wheel' is not installed.
Using legacy 'setup.py install' for pyftpdlib, since package 'wheel' is not installed.

Could also install wheel so pip'll build a wheel for those and store that in the cache for next time?

[leouieda]

Thanks for the tip on caching pip packages!