Describe the task
Spinning up R is very slow, and blocking, it blocks other calls to the same thread in causing timeouts in calls entirely unrelated to anything that uses R

Acceptance Criteria

• CFFDRS R functions exposed as REST endpoints on CFFDRS API
• Feature toggle to switch between CFFDRS using rpy2 and using CFFDRS using API (allows merging code in current state before completing remaining tasks for full switchover)
• Switch from using CFFDRS via rpy2 in PSU API to using CFFDRS via REST interface
• Establish if CFFDRS API is performant enough
• Establish if making adjustments in how we use CFFDRS could make it performant enough (e.g. instead of making hundreds of individual calls, making a single call)
• CFFDRS API can be spun up in docker container on developer computers for local development
• Automated tests for CFFDRS API that can be run locally
• Script(s) for building CFFDRS API image in tools
• Script(s) for deploying CFFDRS API to dev
• Script(s) for deploying CFFDRS API to prod
• Build and deployment process well documented
• CFFDRS API exists as service in dev (e.g. cffdrs on port 80)
• CFFDRS API exists on it's own prod (e.g. cffdrs on port 80)
• CFFDRS API visible only within openshift project
• Feature toggle removed
• rpy2 removed from api/pyproject.toml
• R removed from .github/workflows/integration.yml

Additional context

• See https://github.com/bcgov/wps/blob/task/architecture/README.md#architecture for proposed changes to architecture.
• It is assumed that the implementation will leverage rpy2
• The CFFDRS API won't be frequently deployed, so we don't have to create a pipeline for it. We can rely on manually running automated tests, and manual deployment to dev and prod environments.
• With CFFDRS spun out into it's own micro-service, we can at a later point switch out the underlying calculations to something else (python native, c native, java etc. etc.), without (theoretically) having to touch the PSU api.
• We should probably not be using private functions from CFFDRS R package. (e.g. anything in r that is defined with a period, is actually internal. e.g. .Slopecalc); We should probably be calling fbp <- function(input = NULL, output = "Primary", m = NULL, cores = 1) and disregarding internal functions.