Handle limits for DeleteObjects requests

[evgeniiz321]

if we try to delete a big number of objects at once (e.g. 1001) - s3 gate hangs and a client receives 'Read timeout on endpoint URL: "https://s3.neofs.devenv:8080/yournamehere-93h9ppiq24ny870o-1?delete' instead of a proper error.

2023-10-12T00:19:25.192Z	error	handler/delete.go:254	couldn't delete objects ...
errors": [{"error": "mark object removal via connection pool: remove object via client: delete object on client: rpc error: code = Canceled desc = context canceled"}, {"error": "failed to get node by path: rpc error: code = Canceled desc = context canceled"}
...
023-10-12T00:19:25.193Z	error	handler/util.go:29	call method	{"status": 500, "request_id": "e665a319-686c-425b-a7ab-bac63c4b23ec", "method": "DeleteMultipleObjects", "bucket": "yournamehere-93h9ppiq24ny870o-1", "object": "", "description": "could not write response", "error": "encode xml response: write tcp 192.168.130.82:8080->192.168.130.1:42882: write: broken pipe"

There should be some limits introduced on the amount of objects that can be deleted in a one request. Test test_multi_object_delete_key_limit assumes the limit is 1000, if bigger - it expects to get the 400 status code.

Full logs from the gate - s3_gate.tar.gz