fix: dont panic when schema_config does not exist in the config file

[abacef]

What this PR does / why we need it:
initial startup should not panic

Which issue(s) this PR fixes:
Fixes #16698

Special notes for your reviewer:

Checklist

• Reviewed the CONTRIBUTING.md guide (required)
• Documentation added
• Tests updated
• Title matches the required conventional commits format, see here
  • Note that Promtail is considered to be feature complete, and future development for logs collection will be in Grafana Alloy. As such, feat PRs are unlikely to be accepted unless a case can be made for the feature actually being a bug fix to existing behavior.
• Changes that require user attention or interaction to upgrade are documented in docs/sources/setup/upgrade/_index.md
• If the change is deprecating or removing a configuration option, update the deprecated-config.yaml and deleted-config.yaml files respectively in the tools/deprecated-config-checker directory. Example PR

[CLAassistant]

All committers have signed the CLA.

[trevorwhitney]

I'm confused how is Loki going to work without a schema config? Would you mind sharing the config you're running and how you are storing the data?

[abacef]

The thing I wanted to change it to know why the config is wrong. With this change, the binary outputs this, which helps me debut faster

level=error ts=2025-03-12T18:54:23.284905392Z caller=main.go:70 msg="validating config" err="CONFIG ERROR: invalid schema config: must specify at least one schema configuration"

[chaudum]

I see what the problem is: There's is a panic in the config validation, because it accesses properties of the schema_config even if it is not present. This is definitely a bug and not expected behaviour.

@trevorwhitney Loki still will not start up without a schema_config, because the validation would fail. However it would terminate gracefully.

[chaudum]

I think we should add a comment mentioning that the length of c.SchemaConfig.Configs is already validated in the Validate() function of the SchemaConfig itself.

[abacef]

Ok, I added the comment

[trevorwhitney]

I understand now, lgtm

[abacef]

Apparently I need to lint this code before the lint check can pass. I tried running make format with a new clone of this branch but it does not seem to change any files. Do I have to do any setup to run the linter?