Update to v1.1.1

[pseudoyim]

Notes:

• Skip builds for python<3.8; upstream project began support for only py38 to py310 starting in version 1.1.0.
• Remove tests_to_skip lines for win-32 (we no longer support for this platform).
• As of v1.1.0, scikit-learn requires minimum of numpy-1.17.3. See changelog for v1.1.0.
• Changelog: https://scikit-learn.org/dev/whats_new/v1.1.html#version-1-1-1
• License: https://github.com/scikit-learn/scikit-learn/blob/1.1.1/COPYING
• Update/confirm version pinnings for:
  • joblib
  • numpy
  • scipy
  • setuptools
  • threadpoolctl
  • pytest

[pseudoyim]

Got this error:

ImportError: Your installation of numpy 1.16.6 is out-of-date.
scikit-learn requires numpy >= 1.17.3.

Adding a conda_build_config.yaml to pin to numpy-1.19.

[skupr-anaconda]

There was a missing build for the linux-64 platform. I've reopened the PR