Migrate RDS instance to Terraform

[hackdna]

Toward #2602. Add RDS_ENDPOINT_ADDRESS and RDS_SUPERUSER_PASSWORD variables to config.yaml until migration to TF is complete. Remove RDS_DB_SUBNET_GROUP_NAME and RDS_NAME from config.yaml.

[codecov]

Codecov Report

Merging #3056 into develop will decrease coverage by 1.1%.
The diff coverage is n/a.

@@             Coverage Diff             @@
##           develop    #3056      +/-   ##
===========================================
- Coverage    60.78%   59.67%   -1.11%     
===========================================
  Files          434      434              
  Lines        27879    27041     -838     
  Branches      1275     1275              
===========================================
- Hits         16947    16138     -809     
+ Misses       10932    10903      -29

Impacted Files	Coverage Δ
refinery/data_set_manager/utils.py	84.28% <0%> (-1.83%)	⬇️
refinery/data_set_manager/tests.py	99.28% <0%> (-0.22%)	⬇️
refinery/core/models.py	75.61% <0%> (-0.11%)	⬇️
refinery/core/test_models.py	99.82% <0%> (-0.02%)	⬇️
refinery/tool_manager/models.py	97% <0%> (-0.02%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 406e63a...90a7fa8. Read the comment docs.

[scottx611x]

Cool. Do you think this would be a good point in time to configure and use a failover RDS instance so that this second private subnet will get utilized?

[hackdna]

We have not had an outage yet due to running RDS in a single AZ. Also, if there is a problem in that AZ then the app server instance would be likely affected as well and it does not have a failover configuration, so the site would be down anyway (and this would double the spend on RDS).