fix: handle both string[] and object[] formats in provider-models cache

Category delegation fails when provider-models.json contains model objects
with metadata (id, provider, context, output) instead of plain strings.
Line 196 in model-availability.ts assumes string[] format, causing:
  - Object concatenation: `${providerId}/${modelId}` becomes "ollama/[object Object]"
  - Empty availableModels Set passed to resolveModelPipeline()
  - Error: "Model not configured for category"

This is the root cause of issue #1508 where delegate_task(category='quick')
fails despite direct agent routing (delegate_task(subagent_type='explore'))
working correctly.

Changes:
- model-availability.ts: Add type check to handle both string and object formats
- connected-providers-cache.ts: Update ProviderModelsCache interface to accept both formats
- model-availability.test.ts: Add 4 test cases for object[] format handling

Direct agent routing bypasses fetchAvailableModels() entirely, explaining why
it works while category routing fails. This fix enables category delegation
to work with manually-populated Ollama model caches.

Fixes #1508

src/shared/model-availability.test.ts

@@ -619,7 +619,7 @@ describe("fetchAvailableModels with provider-models cache (whitelist-filtered)",
 		rmSync(tempDir, { recursive: true, force: true })
 	})
 
-	function writeProviderModelsCache(data: { models: Record<string, string[]>; connected: string[] }) {
+	function writeProviderModelsCache(data: { models: Record<string, string[] | any[]>; connected: string[] }) {
 		const cacheDir = join(tempDir, "oh-my-opencode")
 		require("fs").mkdirSync(cacheDir, { recursive: true })
 		writeFileSync(join(cacheDir, "provider-models.json"), JSON.stringify({
@@ -723,6 +723,72 @@ describe("fetchAvailableModels with provider-models cache (whitelist-filtered)",
 		expect(result.has("anthropic/claude-opus-4-5")).toBe(false)
 		expect(result.has("google/gemini-3-pro")).toBe(false)
 	})
+
+	it("should handle object[] format with metadata (Ollama-style)", async () => {
+		writeProviderModelsCache({
+			models: {
+				ollama: [
+					{ id: "ministral-3:14b-32k-agent", provider: "ollama", context: 32768, output: 8192 },
+					{ id: "qwen3-coder:32k-agent", provider: "ollama", context: 32768, output: 8192 }
+				]
+			},
+			connected: ["ollama"]
+		})
+
+		const result = await fetchAvailableModels(undefined, {
+			connectedProviders: ["ollama"]
+		})
+
+		expect(result.size).toBe(2)
+		expect(result.has("ollama/ministral-3:14b-32k-agent")).toBe(true)
+		expect(result.has("ollama/qwen3-coder:32k-agent")).toBe(true)
+	})
+
+	it("should handle mixed string[] and object[] formats across providers", async () => {
+		writeProviderModelsCache({
+			models: {
+				anthropic: ["claude-opus-4-5", "claude-sonnet-4-5"],
+				ollama: [
+					{ id: "ministral-3:14b-32k-agent", provider: "ollama" },
+					{ id: "qwen3-coder:32k-agent", provider: "ollama" }
+				]
+			},
+			connected: ["anthropic", "ollama"]
+		})
+
+		const result = await fetchAvailableModels(undefined, {
+			connectedProviders: ["anthropic", "ollama"]
+		})
+
+		expect(result.size).toBe(4)
+		expect(result.has("anthropic/claude-opus-4-5")).toBe(true)
+		expect(result.has("anthropic/claude-sonnet-4-5")).toBe(true)
+		expect(result.has("ollama/ministral-3:14b-32k-agent")).toBe(true)
+		expect(result.has("ollama/qwen3-coder:32k-agent")).toBe(true)
+	})
+
+	it("should skip invalid entries in object[] format", async () => {
+		writeProviderModelsCache({
+			models: {
+				ollama: [
+					{ id: "valid-model", provider: "ollama" },
+					{ provider: "ollama" },
+					{ id: "", provider: "ollama" },
+					null,
+					"string-model"
+				]
+			},
+			connected: ["ollama"]
+		})
+
+		const result = await fetchAvailableModels(undefined, {
+			connectedProviders: ["ollama"]
+		})
+
+		expect(result.size).toBe(2)
+		expect(result.has("ollama/valid-model")).toBe(true)
+		expect(result.has("ollama/string-model")).toBe(true)
+	})
 })
 
 describe("isModelAvailable", () => {