Add general formatter for chat api, with chat template based (#1728)

Add general formatter for chat api, with chat template based on the model tokenizer

The model_name should be overried at runtime (e.g. chat_api_with_tokenizer_chat_template[model_name=ibm-granite/granite-3.1-2b-instruct])

Currently, we already have HFSystemFormat based formats (formats.models.llava_interleave)
which may be used in the same why. But since the name is very confusing, I
add this format).

Co-authored-by: Elron Bandel <elronbandel@gmail.com>

Author: eladven

prepare/formats/chat_api_with_tokenizer_chat_template.py

@@ -0,0 +1,6 @@
+from unitxt import add_to_catalog
+from unitxt.formats import HFSystemFormat
+
+format = HFSystemFormat(model_name="ibm-granite/granite-3.1-2b-instruct")
+
+add_to_catalog(format, "formats.chat_api_with_tokenizer_chat_template", overwrite=True)

src/unitxt/catalog/formats/chat_api_with_tokenizer_chat_template.json

@@ -0,0 +1,4 @@
+{
+    "__type__": "hf_system_format",
+    "model_name": "ibm-granite/granite-3.1-2b-instruct"
+}