quantize: Categorical features are not supported in block quantization

[NikitaZhidkov]

Hello, could you please add support for categorical features in the utils.quantize method? It's crucial for handling large datasets that include categorical data. Here is the code to reproduce the issue:
`import pandas as pd
from catboost.utils import quantize

data = {
'precipitation': [0, 1, 0, 1],
'temperature': [20.5, 22.3, 18.0, 21.6],
'humidity': [85, 60, 78, 75],
'wind_direction': ['north', 'south', 'east', 'west']
}

df = pd.DataFrame(data)

filename = "weather_data.tsv"
df.to_csv(filename, sep='\t', index=False, header=False)

column_descriptions = {
'Index': ['0', '1', '2', '3'],
'Type': ['Label', 'Num', 'Num', 'Categ'],
'Feature': ['precipitation', 'temperature', 'humidity', 'wind_direction']
}

desc_df = pd.DataFrame(column_descriptions)

desc_filename = "columns_description.tsv"
desc_df.to_csv(desc_filename, sep='\t', index=False, header=False)

pool_quantize = quantize(
data_path=filename,
column_description=desc_filename,
delimiter='\t'
)
`
Returned error: CatBoostError: /Users/zomb-ml-platform-msk/go-agent-21.2.0/pipelines/BuildMaster/catboost.git/catboost/libs/data/load_and_quantize_data.cpp:52: Categorical features are not supported in block quantization