Refactor: enhance graphrag - part 2 (#14972)

### What problem does this PR solve?
1. expose batch_chunk_token_size for configuration
2. retrieve chunks when build subgraph for the doc, not retreive all
docs chunks at the begining
3. get all chunks for a document, used to be hard coded 10000
4. delete not used method run_graphrag

### Type of change

- [x] New Feature (non-breaking change which adds functionality)
- [x] Refactoring

Follow on: #14617

test/testcases/configs.py

@@ -65,6 +65,7 @@ DEFAULT_PARSER_CONFIG = {
             "category",
         ],
         "method": "light",
+        "batch_chunk_token_size": 4096,
     },
     "parent_child": {
         "use_parent_child": False,

test/testcases/test_http_api/test_file_management_within_dataset/test_update_document.py

@@ -387,6 +387,7 @@ DEFAULT_PARSER_CONFIG_FOR_TEST = {
             "category",
         ],
         "method": "light",
+        "batch_chunk_token_size": 4096,
     },
 }
 

test/testcases/test_sdk_api/test_file_management_within_dataset/test_update_document.py

@@ -313,6 +313,7 @@ DEFAULT_PARSER_CONFIG_FOR_TEST = {
             "category",
         ],
         "method": "light",
+        "batch_chunk_token_size": 4096,
     },
 }