chore(logging): enable rotated logging for supervisord to improve debugging

[lau90eth]

What this PR does / Why we need it:

Currently, supervisord activity logs are discarded to /dev/null, making it impossible to debug supervisord itself (e.g., process restarts, config issues, errors).

This PR enables proper rotated logging for the supervisord activity log:

• logfile=/var/log/supervisord/supervisord.log
• Rotation at 50MB with 10 backups
• loglevel=info for useful but not excessive output

Program stdout/stderr remain redirected to container stdout (visible via docker compose logs), no changes there.

Changes:

• Updated [supervisord] section to enable logfile and rotation.
• No breaking changes to program sections.

Verification (tested locally on WSL2 + Docker Desktop):

• docker compose up -d --build → Services start successfully.
• docker compose logs → No loss of program output; supervisord events now loggable (if issues occur).
• Config remains minimal and production-safe.

References:

• Supervisord logging docs: http://supervisord.org/logging.html (recommends logfile_maxbytes and backups for rotation)

Open to adjustments (e.g., different path, size, or per-program logs). Thanks for reviewing!

[cb-heimdall]

🟡 Heimdall Review Status

Requirement	Status	More Info
Reviews	🟡 0/1	Denominator calculation Show calculation 1 if user is bot 0 1 if user is external 0 2 if repo is sensitive 0 From .codeflow.yml 1 Additional review requirements Show calculation Max 0 0 From CODEOWNERS 0 Global minimum 0 Max 1 1 1 if commit is unverified 1 Sum 2

[Dark-Brain07]

Great improvement for debugging! Having rotated supervisord logs is essential for diagnosing restart loops.
The 50MB limit with 10 backups (500MB total) seems reasonable. For operators running archive nodes for extended periods, they might want this documented as a configurable option.
This complements the troubleshooting documentation we're adding for the execution clients.