[BUG]: Group Mira Requests

[mwdchang]

Summary

We are observing quite a decent amount of MIRA fetching concept errors, for example

https://app.staging.terarium.ai/projects/34e7b955-f408-4e8b-be27-9429480ad6aa/model/7508b278-caea-4d35-bb17-097bdd463f07

Which makes a bunch of MIRA queries

http://localhost:8080/api/mira/currie/ido:0000514?project-id=34e7b955-f408-4e8b-be27-9429480ad6aa

This is particularly egregious for stratified models, where many states share the same concepts, and we are sending and failing the same requests dozens of times. Resulting in traffic jam and dozens of error toasters.

Stack trace server side

2024-08-27 00:15:55 [ERROR] An unknown error occurred while MIRA was trying to determine concepts based on curies: ido:0000511 [software.uncharted.terarium.hmiserver.controller.mira.MiraController:394]
2024-08-27 00:15:55 [ERROR] An unknown error occurred while MIRA was trying to determine concepts based on curies: ido:0000514 [software.uncharted.terarium.hmiserver.controller.mira.MiraController:394]
2024-08-27 00:15:55 [ERROR] Servlet.service() for servlet [dispatcherServlet] in context with path [] threw exception [Request processing failed: java.lang.IllegalArgumentException: HttpStatusCode is required] with root cause [org.apache.catalina.core.ContainerBase.[Tomcat].[localhost].[/].[dispatcherServlet]:175]
java.lang.IllegalArgumentException: HttpStatusCode is required
	at org.springframework.util.Assert.notNull(Assert.java:204)
	at org.springframework.http.ProblemDetail.forStatus(ProblemDetail.java:281)
	at org.springframework.web.server.ResponseStatusException.<init>(ResponseStatusException.java:94)
	at org.springframework.web.server.ResponseStatusException.<init>(ResponseStatusException.java:79)
	at org.springframework.web.server.ResponseStatusException.<init>(ResponseStatusException.java:56)
	at software.uncharted.terarium.hmiserver.controller.mira.MiraController.handleMiraFeignException(MiraController.java:402)

Task

• Investigate why we are erroring out, perhaps we didn't set up the environment up properly or maybe the external MIRA service is no longer running
• Consider caching the request so we don't kill the server. See simulation-service which use a promise based FIFO-cache
• Consider batching the requests and consolidate into a single error message instead of several dozens of error messages

[Tom-Szendrey]

also check why we are not getting mira is unavailable error message