fix: Continue summarizer on error and add more logging

[arslanashraf7]

What are the relevant tickets?

https://github.com/mitodl/hq/issues/7123

Description (What does it do?)

• Adds continuation of the command if any content file summarization fails for some reason
• Adds more logging for the command to print. Both success and error messages with details on which content file failed
• Fixes an issue with the Keycloak container on an m4 machine ARM64 arch

Screenshots (if appropriate):

• Desktop screenshots
• Mobile width screenshots

How can this be tested?\

Testing summarize fix

• Follow the same instructions mentioned in feat: content summaries and flashcards generation #2118 to be able to set up the summarization feature.
• To reproduce the error on main branch, you can follow the steps in https://github.com/mitodl/hq/issues/7123 OR do the below
  • Have 4-5 content files set up with some content value
  • Manually reduce the value of [max_tokens](https://github.com/mitodl/mit-learn/blob/main/learning_resources/content_summarizer.py#L238) so that the flashcards response from the LLM would not be complete because of low max_tokens. The JSON response will cut off in the middle and fail the flashcards JSON format validation. See More details in the ticket.
  • When you reduce the max_tokens, make sure you do not set it so low that this fails on the prompt. It should be reasonable enough that the prompt gets through and but their response from LLM breaks.
• When you have reproduced this issue on main, you will notice that all the files after the failing one will not be processed, and the command will stop.
• Once you have reproduced the error on the main branch, now shift to this branch and re-run the summarize command again. This time, you will again hit the flashcards Format error because of a low max_tokens value, but the command will not halt, and it should process all the remaining files.
• At the end, the command should log stats, i.e.
  • Failed file IDs and the error details
  • Succeccful content file Ids

Testing Keycloak fix on ARM64 machine

• Your Keycloak container should run without any errors (Ideal, if you could check it on a container built fresh)

Additional Context

[github-actions]

OpenAPI Changes

Show/hide No detectable change.

[github-actions]

OpenAPI Changes

Show/hide No detectable change.

[arslanashraf7]

My Keycloak container was not starting up because of java errors so I had to do the same thing here that I did with opensearch container. #2069 & #2082

[github-actions]

OpenAPI Changes

Show/hide No detectable change.

[github-actions]

OpenAPI Changes

Show/hide No detectable change.

[github-actions]

OpenAPI Changes

Show/hide No detectable change.

[shanbady]

Worked as expected after syncing to set up initial fixtures properly. As discussed it might be worth adding docs on how to set this up or adding some default fixtures for local envs in a separate pr.