fix(stability): enable Health_Check for the /api/v1/health container check

[TheanLim]

What

Enable Health_Check On in the [SERVICE] block of the two stability templates used by ecs-firelens-stability-tests, keeping the /api/v1/health container health check from #166 unchanged:

• apps/firelens-stability/templates/golden-path-mountebank-fargate-v01-11-2023/fluent-bit.conf
• apps/firelens-stability/templates/s3-fargate-v04-05-2023/fluent-bit.conf

Why

The health check added in #166 uses curl -f http://127.0.0.1:2020/api/v1/health, but the stability configs set HTTP_Server On without Health_Check On. /api/v1/health is only registered as a real endpoint when the Health_Check feature is enabled. Without it:

• On FLB 5.0.x the endpoint 404s (the root / route became exact-match after the upstream
  HTTP server rewrite, fluent/fluent-bit#11538), so every task reports UNHEALTHY.
• On <= 4.x / 1.9.10 the endpoint isn't registered either, but the old Monkey server's catch-all root returns 200 build-info, so the check "passes" without actually reporting health.

Testing

• Confirmed /api/v1/health -> 200 ok on both FLB 1.9.10 and 5.0.9 with Health_Check On.
• Ran stability tests (apps/firelens-stability) using upstream 5.0.9 and all tasks reported healthy.