chore: remove create and read fhir bundle from postgres

[BobanL]

PULL REQUEST

Summary

• Remove data_source variable from core.sql
• Remove postgres from get fhir-data and /save-fhir-data
• Orchestration tests save to azure now
• Remove unused metrics api and uninstall opentelemetry/api

Related Issue

Fixes #2406

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 89.39%. Comparing base (e1385f0) to head (9f074e5).

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3137      +/-   ##
==========================================
+ Coverage   87.03%   89.39%   +2.35%     
==========================================
  Files         221       84     -137     
  Lines       13664     4204    -9460     
  Branches      708      716       +8     
==========================================
- Hits        11892     3758    -8134     
+ Misses       1763      426    -1337     
- Partials        9       20      +11     

Flag	Coverage Δ
ecr-viewer	91.29% <100.00%> (+0.30%)	⬆️
fhir-converter	?
ingestion	?
message-parser	?
message-refiner	?
orchestration	85.67% <ø> (ø)
record-linkage	?
trigger-code-reference	?
validation	?

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
...-viewer/src/app/api/fhir-data/fhir-data-service.ts	94.33% <ø> (-1.32%)	⬇️
...ntainers/ecr-viewer/src/app/api/fhir-data/route.ts	100.00% <ø> (ø)
...ers/ecr-viewer/src/app/api/save-fhir-data/route.ts	76.19% <100.00%> (ø)
...c/app/api/save-fhir-data/save-fhir-data-service.ts	50.00% <100.00%> (+3.52%)	⬆️
containers/ecr-viewer/src/app/api/utils.ts	100.00% <ø> (ø)
.../ecr-viewer/src/app/services/listEcrDataService.ts	74.03% <ø> (ø)

... and 139 files with indirect coverage changes

[mcmcgrath13]

not really a question on this PR, but more generally. When we make a schema change like this, what's the migration plan for users?

[mcmcgrath13]

I'm trying to follow along on what this was doing in the first place. Was this correct? Couldn't we have saved the data to an S3 bucket, but saved the metadata to postgres?

[mcmcgrath13]

also, why don't I see a similar change in the sql server code?

[BobanL]

I believe this was supposed to define where to find the FHIR_DATA for the ecr-viewer (s3 or azure blob), but we only have ever supported fetching the data from a single source.

For your second question, this is only affects postgres because the CORE schema only supports postgres.

@austin-hall-skylight, do you have any thoughts about keeping the data_source variable?