Pass arguments BaseCharmTest.setUpClass

[dshcherb]

Not doing so triggers an incorrect behavior leading to functional test
failures as the application name is not set correctly.

#256
https://review.opendev.org/#/c/712980/1

[codecov-io]

Codecov Report

Merging #257 into master will increase coverage by 0.44%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #257      +/-   ##
==========================================
+ Coverage   20.35%   20.80%   +0.44%     
==========================================
  Files         131      131              
  Lines        7250     7250              
==========================================
+ Hits         1476     1508      +32     
+ Misses       5774     5742      -32     

Impacted Files	Coverage Δ
zaza/openstack/charm_tests/test_utils.py	23.35% <100.00%> (+23.35%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update b821c4c...740bc20. Read the comment docs.

[ajkavanagh]

This is a great catch. It's also a breaking change? It would be good to see some test runs against a few charms to check that there's no unexpected behaviour in other tests that may cause us trouble -- we're 3 days from freeze! Please could we test against, say (in a bastion):

• aodh
• cinder
• nova-compute

and verify there's no weird behaviour.

Note. I may be being over-cautious here, in which case, please feel free to comment!

[dshcherb]

@ajkavanagh The change that introduced the issue was done quite recently: 9cb7876e

So I think it's not a breaking change as far as the older tests go but the ones merged between then and now might be affected.

To find out whether any of them are I did the following:

# None of the lines are prefixed with `+` which means that the lines below are just the lines around new changes that were there before
git log -p 9cb7876e..master | grep -P '\.setUpClass\((\w|\d)+\)'
         super(NeutronOpenvSwitchTest, cls).setUpClass(cls)
         super(GenericPolicydTest, cls).setUpClass(application_name)
         super(BasePolicydSpecialization, cls).setUpClass(application_name)
         super(NeutronGatewayTest, cls).setUpClass(cls)

# Gives more results just for sanity checking and catching multi-line cases
git log -p 9cb7876e..master | grep -P 'setUpClass\('

There is a lesser possibility that some of those test cases began to pass after 9cb7876 was merged.

This change managed to fix it for https://review.opendev.org/#/c/712980/

[ajkavanagh]

@dshcherb okay, I think we are probably pretty safe then. Let's get this merged :)