Support injecting custom config settings files via volume mount #867

What problem does your feature solve?

Currently, to use custom config settings (e.g., custom limits configuration), users must:

1. Add a new settings file to the repository at local/core/etc/config-settings/p25/
2. Build the image using one of these approaches:
   • Push to a branch and let the PR build it
   • Run make build locally
   • Create a new Docker image that inherits from stellar/quickstart and copies in the settings file (avoids rebuilding the base image)
3. Reference the settings file via the --limits option when running the container

This workflow creates friction for users who want to quickly test with custom config settings without building custom images or forking the repository.

What would you like to see?

Support for injecting custom config settings files at runtime via a volume mount path that gets picked up automatically (e.g., /config-settings/).

This would allow users to run something like:

docker run -v ./my-settings:/config-settings stellar/quickstart --limits /config-settings/my-settings.cfg

What alternatives are there?

• Continue with the current workflow (build custom images or fork repo)
• Use --limits unlimited when custom limits aren't strictly necessary
• Add a --limits-json option that accepts raw JSON text, allowing the quickstart image to write it to a file internally (no volume mount needed)
• Inject the settings via stellar container start command, making this a feature accessed through stellar-cli rather than quickstart directly