[feat] Data lineage tracking in `metadata.yaml`

[tscholak]

🧐 Problem Description

Is your feature request related to a specific problem? Please describe it clearly.
For example: "I'm always frustrated when [...]"

💡 Proposed Solution

Describe the solution you would like to see.
Be as specific as possible about how it would work or be implemented.

🔄 Alternatives Considered

Have you considered any alternative solutions or approaches?
If so, please describe them and explain why they might not be ideal.

📈 Potential Benefits

Explain how this feature could benefit Fast-LLM users.
Consider how it might improve performance, usability, scalability, etc.

📝 Additional Context

Add any other context or information that could help us understand the feature request better.
If applicable, provide links to relevant references or examples.

[jlamypoirier]

Very similar to #123, can we merge and close this one?

[tscholak]

Very similar to #123, can we merge and close this one?

Hi @jlamypoirier, @chrish42 is supposed to provide a description here. This is not about adding metadata, this is about tracking dataset lineage information centrally, similar to how wandb tracks training metrics.