Implemented the tool_formatter and tool_extractor for glm4 and Qwen2 tool_format

[mMrBun]

What does this PR do?

Implementation of function call for GLM4,Following the previous implementation method, the PROMPT of the function call is added to the TOOL_FORMAT processing function apply using the PROMPT_FORMAT approach.

• tool_input

tools = [
        {
            "type": "function",
            "function": {
                "name": "get_current_weather",
                "description": "Get the current weather",
                "parameters": {
                    "type": "object",
                    "properties": {
                        "location": {
                            "type": "string",
                            "description": "The city and state, e.g. San Francisco, CA",
                        },
                        "format": {
                            "type": "string",
                            "enum": ["celsius", "fahrenheit"],
                            "description": "The temperature unit to use. Infer this from the users location.",
                        },
                    },
                    "required": ["location", "format"],
                },
            }
        },
        {
            "type": "function",
            "function": {
                "name": "calculate_gpa",
                "description": "Calculate the Grade Point Average (GPA) based on grades and credit hours",
                "parameters": {
                    "type": "object",
                    "properties": {
                        "grades": {"type": "array", "items": {"type": "string"}, "description": "The grades"},
                        "hours": {"type": "array", "items": {"type": "integer"}, "description": "The credit hours"},
                    },
                    "required": ["grades", "hours"],
                },
            },
        }
    ]

• glm4 tokenizer.apply_chat_template

'[gMASK]<sop><|system|>\n你是一个名为 GLM-4 的人工智能助手。你是基于智谱AI训练的语言模型 GLM-4 模型开发的，你的任务是针对用户的问题和要求提供适当的答复和支持。\n\n## get_current_weather\n\n{\n    "name": "get_current_weather",\n    "description": "Get the current weather",\n    "parameters": {\n        "type": "object",\n        "properties": {\n            "location": {\n                "type": "string",\n                "description": "The city and state, e.g. San Francisco, CA"\n            },\n            "format": {\n                "type": "string",\n                "enum": [\n                    "celsius",\n                    "fahrenheit"\n                ],\n                "description": "The temperature unit to use. Infer this from the users location."\n            }\n        },\n        "required": [\n            "location",\n            "format"\n        ]\n    }\n}\n在调用上述函数时，请使用 Json 格式表示调用的参数。\n\n## calculate_gpa\n\n{\n    "name": "calculate_gpa",\n    "description": "Calculate the Grade Point Average (GPA) based on grades and credit hours",\n    "parameters": {\n        "type": "object",\n        "properties": {\n            "grades": {\n                "type": "array",\n                "items": {\n                    "type": "string"\n                },\n                "description": "The grades"\n            },\n            "hours": {\n                "type": "array",\n                "items": {\n                    "type": "integer"\n                },\n                "description": "The credit hours"\n            }\n        },\n        "required": [\n            "grades",\n            "hours"\n        ]\n    }\n}\n在调用上述函数时，请使用 Json 格式表示调用的参数。'

• TOOL_FORMAT

'[gMASK]<sop><|system|>\n你是一个名为 GLM-4 的人工智能助手。你是基于智谱AI训练的语言模型 GLM-4 模型开发的，你的任务是针对用户的问题和要求提供适当的答复和支持，\n\n## get_current_weather\n\n{\n    "name": "get_current_weather",\n    "description": "Get the current weather",\n    "parameters": {\n        "type": "object",\n        "properties": {\n            "location": {\n                "type": "string",\n                "description": "The city and state, e.g. San Francisco, CA"\n            },\n            "format": {\n                "type": "string",\n                "enum": [\n                    "celsius",\n                    "fahrenheit"\n                ],\n                "description": "The temperature unit to use. Infer this from the users location."\n            }\n        },\n        "required": [\n            "location",\n            "format"\n        ]\n    }\n}\n在调用上述函数时，请使用 Json 格式表示调用的参数。\n\n## calculate_gpa\n\n{\n    "name": "calculate_gpa",\n    "description": "Calculate the Grade Point Average (GPA) based on grades and credit hours",\n    "parameters": {\n        "type": "object",\n        "properties": {\n            "grades": {\n                "type": "array",\n                "items": {\n                    "type": "string"\n                },\n                "description": "The grades"\n            },\n            "hours": {\n                "type": "array",\n                "items": {\n                    "type": "integer"\n                },\n                "description": "The credit hours"\n            }\n        },\n        "required": [\n            "grades",\n            "hours"\n        ]\n    }\n}\n在调用上述函数时，请使用 Json 格式表示调用的参数。'

For QWEN2, in the same question
"What's the weather like in San Francisco, Tokyo, and Paris? use Celsius"
the response is
'Action: get_current_weather\nAction Input: {"location": "San Francisco, CA", "format": "celsius"}\nAction: get_current_weather\nAction Input: {"location": "Tokyo, JP", "format": "celsius"}\nAction: get_current_weather\nAction Input: {"location": "Paris, FR", "format": "celsius"}'
Here, the method of identifying tools has been changed from a tuple to a list. For GLM4, I may have conducted fewer tests and haven't encountered cases where multiple tools return results. For now, I will temporarily enclose the tuple in a list in the glm4_tool_extractor function. If I make any new discoveries, I will update the glm4_tool_extractor function accordingly.

Before submitting

• Did you read the contributor guideline?

[hiyouga]

LGTM