feat(i2v): enhance image-to-video documentation and single/dual frame support

- Update README.md to clarify single vs dual frame image support with
automatic model_key selection
- Add detailed explanation of automatic adaptation system for
different image counts
- Include clear distinction between single frame mode (1 image) and
dual frame mode (2 images) with proper model selection

fix(browser-captcha): increase page load timeout from 15s to 60s

- Extend retry range from 15 to 60 attempts for page loading
- Update debug logging to reflect new timeout values
- Improve reliability of captcha service by allowing more time for
page load completion

fix(api): correct I2V API endpoint URL for single frame generation

- Change URL from batchAsyncGenerateVideoStartAndEndImage to
batchAsyncGenerateVideoStartImage for single frame scenarios

fix(model-config): correct model_key for I2V single frame mode

- Fix typo in model_key: remove duplicate 'fl_' in
veo_3_1_i2v_s_fast_fl_landscape
- Implement automatic model_key transformation for single frame mode
by replacing '_fl_' with '_' in model keys
- Add debug logging for model key transformation process

README.md

@@ -117,7 +117,11 @@ python main.py
 | `veo_2_0_t2v_landscape` | 文生视频 | 横屏 |
 
 #### 首尾帧模型 (I2V - Image to Video)
-📸 **支持1-2张图片：首尾帧**
+📸 **支持1-2张图片：1张作为首帧，2张作为首尾帧**
+
+> 💡 **自动适配**：系统会根据图片数量自动选择对应的 model_key
+> - **单帧模式**（1张图）：使用首帧生成视频
+> - **双帧模式**（2张图）：使用首帧+尾帧生成过渡视频
 
 | 模型名称 | 说明| 尺寸 |
 |---------|---------|--------|

src/services/browser_captcha_personal.py

@@ -131,11 +131,11 @@ class BrowserCaptchaService:
         
         # 等待页面加载完成（带重试机制）
         page_loaded = False
-        for retry in range(15):
+        for retry in range(60):
             try:
                 await asyncio.sleep(1)
                 ready_state = await self.resident_tab.evaluate("document.readyState")
-                debug_logger.log_info(f"[BrowserCaptcha] 页面状态: {ready_state} (重试 {retry + 1}/15)")
+                debug_logger.log_info(f"[BrowserCaptcha] 页面状态: {ready_state} (重试 {retry + 1}/60)")
                 if ready_state == "complete":
                     page_loaded = True
                     break
@@ -391,7 +391,7 @@ class BrowserCaptchaService:
             
             # 等待页面加载完成
             page_loaded = False
-            for retry in range(15):
+            for retry in range(60):
                 try:
                     await asyncio.sleep(1)
                     ready_state = await tab.evaluate("document.readyState")
@@ -402,7 +402,7 @@ class BrowserCaptchaService:
                     debug_logger.log_warning(f"[BrowserCaptcha] 标签页连接丢失: {e}")
                     return None
                 except Exception as e:
-                    debug_logger.log_warning(f"[BrowserCaptcha] 等待页面异常: {e}，重试 {retry + 1}/15...")
+                    debug_logger.log_warning(f"[BrowserCaptcha] 等待页面异常: {e}，重试 {retry + 1}/60...")
                     await asyncio.sleep(1)
             
             if not page_loaded:

src/services/flow_client.py

@@ -572,7 +572,7 @@ class FlowClient:
         Returns:
             同 generate_video_text
         """
-        url = f"{self.api_base_url}/video:batchAsyncGenerateVideoStartAndEndImage"
+        url = f"{self.api_base_url}/video:batchAsyncGenerateVideoStartImage"
 
         # 获取 reCAPTCHA token
         recaptcha_token = await self._get_recaptcha_token(project_id) or ""

src/services/generation_handler.py

@@ -118,7 +118,7 @@ MODEL_CONFIG = {
     "veo_3_1_i2v_s_fast_fl_landscape": {
         "type": "video",
         "video_type": "i2v",
-        "model_key": "veo_3_1_i2v_s_fast_landscape_fl_ultra_relaxed",
+        "model_key": "veo_3_1_i2v_s_fast_fl_ultra_relaxed",
         "aspect_ratio": "VIDEO_ASPECT_RATIO_LANDSCAPE",
         "supports_images": True,
         "min_images": 1,
@@ -594,12 +594,16 @@ class GenerationHandler:
                         user_paygate_tier=token.user_paygate_tier or "PAYGATE_TIER_ONE"
                     )
                 else:
-                    # 只有首帧
+                    # 只有首帧 - 需要将 model_key 中的 _fl_ 替换为 _
+                    # 例如: veo_3_1_i2v_s_fast_fl_ultra_relaxed -> veo_3_1_i2v_s_fast_ultra_relaxed
+                    #       veo_3_1_i2v_s_fast_portrait_fl_ultra_relaxed -> veo_3_1_i2v_s_fast_portrait_ultra_relaxed
+                    actual_model_key = model_config["model_key"].replace("_fl_", "_")
+                    debug_logger.log_info(f"[I2V] 单帧模式，model_key: {model_config['model_key']} -> {actual_model_key}")
                     result = await self.flow_client.generate_video_start_image(
                         at=token.at,
                         project_id=project_id,
                         prompt=prompt,
-                        model_key=model_config["model_key"],
+                        model_key=actual_model_key,
                         aspect_ratio=model_config["aspect_ratio"],
                         start_media_id=start_media_id,
                         user_paygate_tier=token.user_paygate_tier or "PAYGATE_TIER_ONE"