Bump Lambda to 4096 MB and cap Node.js heap to prevent OOM

[hiroshinishio]

Summary

• Bump Lambda memory from 3072 to 4096 MB to prevent frequent OOM kills during Jest + MongoMemoryServer runs
• Read memory limit from AWS runtime env var (AWS_LAMBDA_FUNCTION_MEMORY_SIZE) instead of hardcoding, so infra changes auto-propagate
• Cap Node.js V8 heap via NODE_OPTIONS=--max-old-space-size (Lambda memory minus 1.5 GB reserved for Python + mongod + OS), so Jest fails with a catchable JS error instead of killing the Lambda
• Update LGTM workflow: user stages files before saying "lgtm", remove separate git add step

Social Media Post (GitAuto)

Cap Node.js heap to prevent Lambda OOM during Jest runs

• Lambda memory bumped from 3 GB to 4 GB
• Node.js heap capped so Jest crashes gracefully instead of killing the Lambda
• Memory limit now reads from AWS runtime env var instead of being hardcoded

Social Media Post (Wes)

Our Lambda kept dying mid-Jest run. MongoMemoryServer plus a big import chain ate all 3 GB. Bumped to 4 GB and capped Node.js heap so it throws a catchable error instead of silently killing the whole process. The real lesson: if your subprocess can eat unbounded memory, cap it at the OS level, don't trust application-level limits alone.