Refs #23663.

This helps with testability of management commands.

Thanks to trac username daveoncode for the report and to
Tim Graham and Claude Paroz for the reviews.

Refs #23685.

Fixed a regression introduced by 28efafa2.
Thanks Michal Čihař for the report and initial patch, and
Collin Anderson and Tim Graham for the reviews.

This commit reverts 67d7da5f.

The previous fix changed the environment globally, which meant
that any call to `call_command(no_color=True)` prevented further
`call_command` with color.

This fix still relies on the environment because it's currently the only
way to reach WSGIRequestHandler, but it's now limited to the `runserver`
command. This seems an acceptable compromise considering `runserver` runs
indefinitely.

Thanks Tim Graham for the review.

This fixes a regression introduced by 2a31d009.
Thanks tony-zhu for the report.

Thanks Baptiste Mispelon for the report.

Thanks Paul Dejean for the report.

Thanks Naddiseo for reporting.

Thanks Aymeric Augustin for review.

This reverts commit 8d6e1afe.
call_command now always set skip_checks to True (tested in
user_commands).

Thanks Loic Bistuer for the report/review and Tim Graham for the
review.

`cache.tests.TestEtagWithAdmin` loaded views from the `admin_views` test
package. This is problematic because when the `cache` test package is
run in isolation, `admin_views` isn't in INSTALLED_APPS, and therefore
loading its models isn't allowed since the app loading refactor.

Overriding execute() bypasses the system check framework among other things.

Fixed a typo in urlencode documentation.

Thanks Ilya Baryshev for the report and Tim Graham for the review.

Checked database destruction for ORA-29857, which happens when objects of
some types are left over in an old test database (notably, such objects
are created by the GIS backend). When this happens, we can resolve the
issue by dropping the test-user (if we are allowed to).

An alternative approach, to just switch the order of creation (so that,
if the test user exists and can be dropped, it is always dropped before
the tablespace) was considered; but since the user creation depends on
the tablespace, this would necessitate separating the dropping of the user
from its creation -- a change I am reluctant to make.

Thanks Josh Smeaton and Tim Graham for reviews.

Thanks jamesbeith for the report.

The "check" name is a reserved word used by Django's check framework,
and cannot be redefined as something else other than a method, or the check
framework will raise an error.

This change amends the django.core.checks.model_check.check_all_models()
function, so that it verifies that a model instance's attribute "check"
is actually a method. This new check is assigned the id "models.E020".

Thanks Daniele Procida for suggested wording.