Move log-collection library into collector-contrib codebase

[djaglowski]

This is one of the major steps defined in #8850.

This PR copies all packages from opentelemetry-log-collection
into a new pkg/stanza module and updates paths such that
the code will compile and tests will pass. This codebase represents
the exact contents of the log-collection packages at v0.29.0,
with the following exceptions:

• Updated module and package paths
• Added a Makefile and doc.go
• Updated versions.yaml
• Omitted internal/tools module
• Removed a couple extraneous docs
• Removed mockery, as it was barely used
• Resolves several minor gosec issues
• Resolved several minor lint issues, eg:
  • misspellings in docs
  • shadow declarations of err
  • improperly cased initialisms
• Destuttered a massive number of struct names (also for lint)
  • eg file.Input instead of file.FileInput

It does NOT change usages of the log-collection library by components
in the collector. This will follow in a subsequent PR.

[djaglowski]

This PR is massive. It would arguably be an ok approach if it were a straight copy/paste. However, it evolved into a much more diverse set of changes due to differences in lint/gosec rules. I believe a more traceable approach is needed.

Options would seem to be roughly: 1) break this PR up into smaller PRs applying lint as needed but in smaller batches, 2) copy/paste into contrib, but exclude from CI until lint updates are made, or 3) update lint standards on log-collection and apply changes in place first, then bring over, or . In all cases, changes need to be made in logical steps so that they can be reasonably followed.

Option 1 would still obfuscate changes, but it is at least straightforward and would avoid the massive lint update that would be necessary to make the other two options pass CI. Unless there are objections, I'll establish the module, bring over docs, then small core packages like entry, operator, and pipeline, then finally batch up the operators as seems appropriate.

[djaglowski]

Closing in favor of incremental approach. #9328 is the first incremental PR.