fix: better error for failed run section

[mikemccracken]

run section output can be long and it can be a pain to find which image failed to build in multiple-image situations.

This just adds the obvious information to the error you see at the end.

now it looks like this:

Error: run commands failed for image "layer1" in "/home/embrane/ssd/stacker/stackertest-test_basic_workings.XFZkds/stacker.yaml": execute failed: exit status 1

✅ By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

[raharper]

Looks like a nice improvement. Thanks for looking at this.
+1 on this after "workflow" is happy and squashed commits.

[hallyn]

Weird, I don't seem to have an option to re-run the tests.

[mikemccracken]

rebased to get the convert.bats fix from #753

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 52.95%. Comparing base (9104248) to head (0fa7d1d).
⚠️ Report is 16 commits behind head on main.

Additional details and impacted files

@@             Coverage Diff             @@
##             main     #752       +/-   ##
===========================================
- Coverage   63.58%   52.95%   -10.64%     
===========================================
  Files          57       59        +2     
  Lines        5083     6538     +1455     
===========================================
+ Hits         3232     3462      +230     
- Misses       1184     2431     +1247     
+ Partials      667      645       -22     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[hallyn]

Sorry, can you explain why the second commit is part of this pr?

[mikemccracken]

Sorry, can you explain why the second commit is part of this pr?

I can, yes - in a previous attempt I made to fix the failing convert test (which had been blocking this PR, until Tanmay fixed it correctly in a later PR) - I bumped the elasticsearch version too high and hit a version of their dockerfile that wrote to WORKDIR without creating it. See, in v8.17.10, they do a RUN mkdir /usr/share/elasticsearch first: https://github.com/elastic/dockerfiles/blob/3861498adce22926e852b1bbec340f159147a47f/elasticsearch/Dockerfile#L45

but in later versions they do just the WORKDIR: https://github.com/elastic/dockerfiles/blob/v9.2.4/elasticsearch/Dockerfile#L30
which is correct because Docker promises to mkdir your WORKDIR for you.

So although Tanmay's fix means we don't need this for passing CI, I left it in because it improves docker build compat.