feat(canvas): add video-prompt node and enhance video generation support

- Introduced a new node type "video-prompt" for AI video generation, including its integration into the canvas command palette and node template picker.
- Updated connection validation to allow connections from text nodes to video-prompt and from video-prompt to ai-video nodes.
- Enhanced error handling and messaging for video generation failures, including specific cases for provider issues.
- Added tests to validate new video-prompt functionality and connection policies.
- Updated localization files to include new labels and prompts for video-prompt and ai-video nodes.

tests/lib/ai-video-models.test.ts

@@ -0,0 +1,73 @@
+import { describe, expect, it } from "vitest";
+
+import {
+  DEFAULT_VIDEO_MODEL_ID,
+  VIDEO_MODELS,
+  getAvailableVideoModels,
+  getVideoModel,
+  isVideoModelId,
+} from "@/lib/ai-video-models";
+
+describe("ai video models registry", () => {
+  it("contains all planned MVP models", () => {
+    expect(Object.keys(VIDEO_MODELS)).toHaveLength(5);
+    expect(Object.keys(VIDEO_MODELS)).toEqual([
+      "wan-2-2-480p",
+      "wan-2-2-720p",
+      "kling-std-2-1",
+      "seedance-pro-1080p",
+      "kling-pro-2-6",
+    ]);
+    expect(isVideoModelId(DEFAULT_VIDEO_MODEL_ID)).toBe(true);
+  });
+
+  it("keeps credit costs consistent for 5s and 10s durations", () => {
+    for (const model of Object.values(VIDEO_MODELS)) {
+      expect(model.creditCost[5]).toBeGreaterThan(0);
+      expect(model.creditCost[10]).toBeGreaterThan(0);
+      expect(model.creditCost[10]).toBe(model.creditCost[5] * 2);
+    }
+  });
+
+  it("filters available models by tier", () => {
+    expect(getAvailableVideoModels("free").map((model) => model.id)).toEqual([
+      "wan-2-2-480p",
+      "wan-2-2-720p",
+    ]);
+    expect(getAvailableVideoModels("starter").map((model) => model.id)).toEqual([
+      "wan-2-2-480p",
+      "wan-2-2-720p",
+      "kling-std-2-1",
+      "seedance-pro-1080p",
+    ]);
+    expect(getAvailableVideoModels("pro").map((model) => model.id)).toEqual([
+      "wan-2-2-480p",
+      "wan-2-2-720p",
+      "kling-std-2-1",
+      "seedance-pro-1080p",
+      "kling-pro-2-6",
+    ]);
+  });
+
+  it("supports lookup and model id guards", () => {
+    expect(isVideoModelId("wan-2-2-480p")).toBe(true);
+    expect(isVideoModelId("not-a-model")).toBe(false);
+
+    const validModel = getVideoModel("wan-2-2-720p");
+    expect(validModel?.label).toBe("WAN 2.2 720p");
+
+    expect(getVideoModel("not-a-model")).toBeUndefined();
+  });
+
+  it("stores model-specific status endpoints for polling", () => {
+    expect(getVideoModel("wan-2-2-480p")?.statusEndpointPath).toBe(
+      "/v1/ai/text-to-video/wan-2-5-t2v-720p/{task-id}",
+    );
+    expect(getVideoModel("seedance-pro-1080p")?.statusEndpointPath).toBe(
+      "/v1/ai/video/seedance-1-5-pro-1080p/{task-id}",
+    );
+    expect(getVideoModel("kling-std-2-1")?.statusEndpointPath).toBe(
+      "/v1/ai/image-to-video/kling-v2-1/{task-id}",
+    );
+  });
+});

tests/lib/video-poll-logging.test.ts

@@ -0,0 +1,22 @@
+import { describe, expect, it } from "vitest";
+
+import {
+  shouldLogVideoPollAttempt,
+  shouldLogVideoPollResult,
+} from "@/lib/video-poll-logging";
+
+describe("video poll logging", () => {
+  it("logs only the first and every fifth in-progress attempt", () => {
+    expect(shouldLogVideoPollAttempt(1)).toBe(true);
+    expect(shouldLogVideoPollAttempt(2)).toBe(false);
+    expect(shouldLogVideoPollAttempt(5)).toBe(true);
+    expect(shouldLogVideoPollAttempt(6)).toBe(false);
+  });
+
+  it("always logs terminal poll results", () => {
+    expect(shouldLogVideoPollResult(2, "IN_PROGRESS")).toBe(false);
+    expect(shouldLogVideoPollResult(5, "IN_PROGRESS")).toBe(true);
+    expect(shouldLogVideoPollResult(17, "COMPLETED")).toBe(true);
+    expect(shouldLogVideoPollResult(3, "FAILED")).toBe(true);
+  });
+});