[CT-2707] Derived metrics throw an error when using the run based semantic manifest.

[Jstein77]

Is this a new bug in dbt-core?

• I believe this is a new bug in dbt-core
• I have searched the existing issues, and I could not find an existing issue for this bug

Current Behavior

I'm getting this error when querying a derived metric: ins.ValueError: Unknown metric: MetricReference(element_name='order_gross_profit') in element set. I noticed that the measures list in the semantic manifest is empty, which is likely causing the error:

@QMalcolm per our conversation today in standup we need to update the way we populate measures.

 {
            "name": "order_gross_profit",
            "description": "Gross profit from each order.",
            "type": "derived",
            "type_params": {
                "measure": null,
                "measures": [],
                "numerator": null,
                "denominator": null,
                "expr": "revenue - cost",
                "window": null,
                "grain_to_date": null,
                "metrics": [
                    {
                        "name": "food_order_amount",
                        "filter": null,
                        "alias": "revenue",
                        "offset_window": null,
                        "offset_to_grain": null
                    },
                    {
                        "name": "order_amount",
                        "filter": null,
                        "alias": "cost",
                        "offset_window": null,
                        "offset_to_grain": null
                    }
                ]
            },
            "filter": null,
            "metadata": null
        },

Expected Behavior

I should be able to query derived metrics from the run based manifest.

Steps To Reproduce

1. Create a derived metric like the one in jaffle-sl-template: https://github.com/dbt-labs/jaffle-sl-template/blob/d7ce5ba300bbe809a9d36578912c36d1379175f5/models/metrics/example_metrics.yaml#L48
2. Kick of a run in dbt cloud
3. Run semantic_layer.query()

Relevant log output

No response

Environment

- OS:
- Python:
- dbt:

Which database adapter are you using with dbt?

No response

Additional Context

No response

[QMalcolm]

So what we haven't been doing is properly populating the measures attribute of metric type params (additionally measures on metric type params because input_measures once work for #7898 is merged). The input_measures can only get populated once all metrics have gone through initial parsing. This is because of derived metrics. If a derived metric depends on a metric that hasn't been parsed yet, then it won't be able to properly populate the input_measures. An example of how the logic for setting input_measures plays out can be found in dbt-semantic-interfaces' add_input_metric_measures.py transformation rule. I imagine the logic we used would be quite similar. If so we should longer term coordinate with the semantic layer team to abstract this logic in dbt-semantic-interfaces to make it more accessible.