fix: documentation inconsistencies (#6)

* Fix model name inconsistency in documentation

* Fix callback event documentation inaccuracies

Correct the documentation to accurately reflect which callback events
are supported by each streaming type:
- Object: on_start, on_complete, on_update
- List: on_start, on_complete, on_append
- String: on_start, on_complete, on_append
- Atom: on_start, on_complete

Previously, documentation incorrectly suggested all types support
on_append, on_update, and on_complete uniformly.

* fix: preserve the original sentence

Author: rajephon

docs/api/overview.md

@@ -16,7 +16,7 @@ The parser module contains streaming-aware data types and the core parser for pr
 
 ### Key Features
 
-- **Event Callbacks**: All streaming types support `on_start`, `on_append`, and `on_complete` callbacks
+- **Event Callbacks**: All streaming types support `on_start` and `on_complete` callbacks. Additionally, `Object` supports `on_update`, while `List` and `String` support `on_append`
 - **Type Safety**: Full type hints and generic support for compile-time checking
 - **Pydantic Integration**: Convert streaming models to Pydantic models via `to_pydantic()`
 

docs/api/parser.md

@@ -117,7 +117,7 @@ response_format = BlogPost.to_pydantic()
 
 client = openai.OpenAI()
 with client.chat.completions.stream(
-    model="gpt-4o-mini",
+    model="gpt-5-mini",
     messages=[{"role": "user", "content": "Write a blog post"}],
     response_format=response_format,
 ) as stream:
@@ -130,7 +130,7 @@ with client.chat.completions.stream(
 
 ## Event System
 
-All streaming types support three main events:
+All streaming types support common events, with type-specific additional events:
 
 ### on_start()
 Called when streaming begins for a value:
@@ -142,12 +142,25 @@ def on_title_start():
 ```
 
 ### on_append()
-Called as new data is appended:
+Called as new data is appended (supported by `String` and `List` types):
 
 ```python
-@response.content.on_append
+@response.content.on_append  # String type
 def on_content_chunk(chunk: str):
     print(f"New content: {chunk}")
+
+@response.items.on_append  # List type
+def on_item_append(item: ld.String, index: int):
+    print(f"New item at index {index}")
+```
+
+### on_update()
+Called when an object is updated (supported by `Object` type only):
+
+```python
+@response.on_update  # Object type
+def on_object_update(data: dict):
+    print(f"Object updated: {data}")
 ```
 
 ### on_complete()

docs/index.md

@@ -10,7 +10,7 @@ LangDiff provides intelligent partial parsing with granular, type-safe events as
 
 ### Streaming Parsing
 - Define schemas for streaming structured outputs using Pydantic-style models
-- Receive granular, type-safe callbacks (`on_append`, `on_update`, `on_complete`) as tokens stream in
+- Receive granular, type-safe callbacks (`on_append`, `on_update`, `on_complete`) as tokens stream in.
 - Derive Pydantic models from LangDiff models for seamless interop with existing libraries and SDKs like OpenAI SDK
 
 ### Change Tracking
@@ -69,7 +69,7 @@ def on_section_append(section: ld.String, index: int):
 # Stream from OpenAI
 client = openai.OpenAI()
 with client.chat.completions.stream(
-    model="gpt-4o-mini",
+    model="gpt-5-mini",
     messages=[{"role": "user", "content": "Write a short article about Python"}],
     response_format=ArticleResponse.to_pydantic(),
 ) as stream: