refactor(appkit): unify eval CLI to a single --warehouse-id flag

The eval CLI carried two flags for one physical SQL warehouse: --warehouse-id
(MLflow assessment writes to UC-backed traces) and --warehouse (managed-dataset
reads). Collapse to --warehouse-id, feeding both paths; the MLflow write path
keeps its MLFLOW_TRACING_SQL_WAREHOUSE_ID env fallback.

Also apply oxfmt to eval files left unformatted by an earlier rebase.

Signed-off-by: MarioCadenas <MarioCadenas@users.noreply.github.com>

Author: MarioCadenas

packages/appkit/src/connectors/mlflow/auth.ts

@@ -1,4 +1,7 @@
-import { createWorkspaceClient, type WorkspaceClient } from "../../workspace-client";
+import {
+  createWorkspaceClient,
+  type WorkspaceClient,
+} from "../../workspace-client";
 
 /** Resolved Databricks host + bearer token for the eval runner's REST calls. */
 export interface DatabricksAuth {

packages/appkit/src/evals/run-evals.ts

@@ -1,4 +1,5 @@
 import { pathToFileURL } from "node:url";
+
 import { MlflowClient } from "../connectors/mlflow";
 import type { WorkspaceClient } from "../workspace-client";
 import { type DatasetRow, readEvalDataset } from "./dataset";
@@ -496,7 +497,15 @@ export async function runEvalsInDir(
         ...options,
         timeoutMs: options.timeoutMs ?? configs.get(d.agent)?.timeoutMs,
       };
-      await runDiscovered(d, index, total, runId, fileOptions, emit, fileResults);
+      await runDiscovered(
+        d,
+        index,
+        total,
+        runId,
+        fileOptions,
+        emit,
+        fileResults,
+      );
       return fileResults;
     });
     const results = perFile.flat();

packages/appkit/src/evals/tests/run-eval.test.ts

@@ -1,4 +1,5 @@
 import { describe, expect, test, vi } from "vitest";
+
 import { defineEval } from "../define-eval";
 import { includes } from "../matchers";
 import { runEval } from "../run-eval";

packages/shared/src/cli/commands/agent/eval.ts

@@ -1,4 +1,5 @@
 import fs from "node:fs";
+
 import { Command, Option } from "commander";
 
 interface EvalRunSummary {
@@ -109,7 +110,6 @@ interface EvalOptions {
   judgeModel?: string;
   concurrency?: number;
   warehouseId?: string;
-  warehouse?: string;
   timeout?: string;
   retries?: string;
   minPassRate?: string;
@@ -243,8 +243,9 @@ async function runAgentEval(
   const auth: Auth = (await runner.resolveDatabricksAuth(credentials)) ?? {};
 
   // Managed-dataset reads: a workspace client (same profile/host/token) + a SQL
-  // warehouse. Only needed by evals that declare `dataset`.
-  const warehouseId = opts.warehouse ?? process.env.DATABRICKS_WAREHOUSE_ID;
+  // warehouse. Only needed by evals that declare `dataset`. Shares the single
+  // --warehouse-id flag with the MLflow assessment-write path.
+  const warehouseId = opts.warehouseId ?? process.env.DATABRICKS_WAREHOUSE_ID;
   const workspaceClient = runner.resolveWorkspaceClient(credentials);
 
   // Runner-level default per-eval timeout (ms). A per-eval `timeoutMs` wins.
@@ -385,11 +386,7 @@ export const agentEvalCommand = new Command("eval")
   )
   .option(
     "--warehouse-id <id>",
-    "SQL warehouse id for writing assessments to UC-backed experiments (default: MLFLOW_TRACING_SQL_WAREHOUSE_ID or DATABRICKS_WAREHOUSE_ID)",
-  )
-  .option(
-    "--warehouse <id>",
-    "SQL warehouse id for reading managed evaluation datasets (default: DATABRICKS_WAREHOUSE_ID)",
+    "SQL warehouse id for reading managed eval datasets and writing assessments to UC-backed experiments (default: DATABRICKS_WAREHOUSE_ID, or MLFLOW_TRACING_SQL_WAREHOUSE_ID for assessments)",
   )
   .option(
     "--judge-model <endpoint>",