Allow update of zcollection global and variable metadata

[robin-cls]

Hello, I have a suggestion for improving a collection management. For now, it seems impossible to update global and variable metadata once a collection is created. It would be a great addition if we could update or edit the attributes of a variable or the global attributes

Code for collection creation

from __future__ import annotations

from typing import Iterator
import datetime
import pprint
import dask_jobqueue
import os
import dask.distributed
import fsspec
import numpy
import dask

import zcollection
import zcollection.tests.data

cluster = dask.distributed.LocalCluster(processes=False)
client = dask.distributed.Client(cluster)

zds = next(zcollection.tests.data.create_test_dataset_with_fillvalue())

fs = fsspec.filesystem('memory')

partition_handler = zcollection.partitioning.Date(('time', ), resolution='M')
collection: zcollection.Collection = zcollection.create_collection(
    'time', zds, partition_handler, '/my_collection', filesystem=fs)

collection.insert(zds)

Below is a snippet that I think can help illustrate what we want to achieve

print(f'Global attributes before update: {collection.load().attrs}')
print(f'Variable attribute before update: {collection.load()["var1"].attrs}')
>> Global attributes before update: (Attribute('attr', 1),)
>> Variable attribute before update: (Attribute('attr', 1),)

# These would be the new methods. Eventually they could be called from the
# collection instance
update_global_attributes(collection, [Attribute('attr', 2), Attribute('foo', 'bar')])
update_variable_attributes(collection, [Attribute('attr', 3), Attribute('fooz', 'baz')])

print(f'Global attributes after update: {collection.load().attrs}')
print(f'Variable attribute after update: {collection.load()["var1"].attrs}')
>> Global attributes after update: (Attribute('attr', 2), Attribute('foo', 'bar'))
>> Variable attribute after update: (Attribute('attr', 3), Attribute('fooz', 'baz'))