feat: add front-end tests

[wazelin]

FUN-965

This PR enables client-side test and includes them into the report publication.

See https://github.com/oat-sa/tao-community/actions/runs/5544605113/jobs/10122308704#step:11:444 and #517 (comment).

Execution Time (2023-07-13 14:55:02 UTC-0)
loading tasks                   1.1s  ▇ 2%
qunit:qtiitempcitest            4.2s  ▇▇ 6%
qunit:taotest                  11.6s  ▇▇▇▇▇▇ 17%
qunit:taoclientdiagnostictest   3.2s  ▇▇ 5%
qunit:taodeliveryrdf           833ms  ▇ 1%
qunit:taoitemstest             873ms  ▇ 1%
qunit:taomediamanagertest       2.1s  ▇ 3%
qunit:taoqtiitemtest           19.5s  ▇▇▇▇▇▇▇▇▇ 28%
qunit:taoqtitest                6.7s  ▇▇▇▇ 10%
qunit:taoqtitestpreviewer      15.7s  ▇▇▇▇▇▇▇▇ 23%
qunit:taotaskqueuetest          1.6s  ▇ 2%
Total 1m 9.1s

[github-actions]

Back-end summary PHP 8.1

💯 Total	✅ Passed	⏭️ Skipped	❌ Failed
1816	1803	13	0

[github-actions]

Back-end summary PHP 8.0

💯 Total	✅ Passed	⏭️ Skipped	❌ Failed
3516	3494	22	0

[github-actions]

Back-end summary PHP 7.4

🔗 Coverage report

💯 Total	✅ Passed	⏭️ Skipped	❌ Failed
3516	3494	22	0

[github-actions]

Front-end summary Node 18

💯 Total	✅ Passed	⏭️ Skipped	❌ Failed
1814	1814	0	0

[kilatib]

LGTM

[oatymart]

Great job on getting FE tests to fully run, and improving the pipeline again!

If you want to stop at this version I don't mind, but if we would spend an extra day or so we could modify the grunt files in tao-core so it can get the good puppeteer args for this environment in a regular way?

[oatymart]

There must be a saner way to set some args, than editing a file on the fly!

[wazelin]

For sure :)

In fact, I noticed only later that the flags I added probably do not make much of a difference. Let me try to execute the workflow without this ugly line.

[wazelin]

Yep.

The run was successful without these flags as well. So, while they could be adding to stability of the tests, I guess we can live without this ugly hack.