Cleanup environment installation steps

CONTRIBUTING.md

@@ -85,12 +85,18 @@ environments. Our recommended setup for contributing is:
   ``cd metpy``
 * Connect your repository to the upstream (main project).
   ``git remote add unidata https://github.com/unidata/metpy.git``
-* Create the development environment by running
-  ``conda create -n devel -c conda-forge --file ci/requirements.txt --file ci/extra_requirements.txt
-  --file ci/test_requirements.txt``.
-  This will install all the necessary packages for development.
-* Activate our new development environment by running ``conda activate devel``
-* Make an editable install of MetPy by running ``pip install -e .``
+* Create a new conda environment for us to configure, and give it a name with
+  ``conda create -n devel``
+* **IMPORTANT**: activate this new environment with ``conda activate devel``.
+* Configure this environment so that we can reach
+  [conda-forge](https://conda-forge.org/feedstock-outputs/) for our packages with
+  ``conda config --env --add channels conda-forge --add channels conda-forge/labels/testing``
+* Install the necessary packages from conda-forge with 
+  ``conda install --file ci/requirements.txt --file ci/extra_requirements.txt --file ci/test_requirements.txt``.
+* Finally, make an editable install of MetPy by running ``pip install -e .``
+
+Note sections on [documentation](#documentation) and [code style](#code-style) below,
+where you may need to install a few more packages into your new environment.
 
 Now you're all set! You have an environment called ``devel`` that you can work in. You'll need
 to make sure to activate that environment next time you want to use it after closing the
@@ -151,7 +157,9 @@ You can write examples in the documentation if they are simple concepts to demon
 your feature is more complex, consider adding to the examples or tutorials for MetPy.
 
 You can build the documentation locally to see how your changes will look.
-
+After setting up your [development environment](#setting-up-your-development-environment) above,
+use ``conda install --file ci/doc_requirements.txt`` to install required packages to build our documentation.
+Then, from within your ``devel`` environment:
 * Navigate to the docs folder ``cd docs``
 * Remove any old builds and build the current docs ``make clean html``
 * Open ``docs/build/html/index.html`` and see your changes!
@@ -262,9 +270,10 @@ While the authors are no fans of blind adherence to style and so-called project
 that go through and correct code style, MetPy has adopted this style from the outset.
 Therefore, it makes sense to enforce this style as code is added to keep everything clean and
 uniform. To this end, part of the automated testing for MetPy checks style. To check style
-locally within the source directory you can use the ``flake8`` tool. Running it
-from the root of the source directory is as easy as running ``pytest --flake8`` in the base
-of the repository.
+locally within the source directory you can use the ``flake8`` tool.
+After setting up your [development environment](#setting-up-your-development-environment) above,
+install the code style tools we use with ``conda install --file ci/linting_requirements.txt``.
+Checking your code style is then as easy as running ``pytest --flake8`` in the base of the repository.
 
 You can also just submit your PR and the kind robots will comment on all style violations as
 well. It can be a pain to make sure you have the right number of spaces around things, imports