unload_table fails with KMS encypted s3 bucket, option KMS_KEY_ID misses

### Describe the bug

When the S3 bucket requires encryption

The error received is:

S3ServiceException:User: arn:aws:sts:::assumed-role/redshift-s3-role/RedshiftIamRoleSession is not authorized to perform: s3:PutObject on resource: "arn:aws:s3:::bucketname..."

### Steps to reproduce

Create a bucket and a KMS key
Create bucket policy that does not allow PutObject unless the are encrypted

"Sid": "DenyUnEncryptedObjectUploads",
"Effect": "Deny",
"Principal": {
"AWS": "*"
},
"Action": "s3:PutObject",
"Resource": "arn:aws:s3:::bucketname/*",
"Condition": {
"Null": {
"s3:x-amz-server-side-encryption": "true"
}
}
Try use the unload_table macro

### Expected results

It should succeed as long as you provide KMS_KEY_ID but that missed

### Actual results

The error mentioned above occurs

### System information

- [ ] postgres
- [x] redshift
- [ ] bigquery
- [ ] snowflake
- [ ] other (specify: ____________)

I can create a PR for this. Let my know :-)

I was able to work around this creating my own wrapper:

{% macro my_unload_table(
schema,
table,
s3_path,
iam_role=None,
aws_key=None,
aws_secret=None,
aws_region=None,
aws_token=None,
manifest=False,
header=False,
format=None,
delimiter=",",
null_as="",
max_file_size='6 GB',
escape=True,
compression=None,
add_quotes=False,
encrypted=False,
overwrite=False,
cleanpath=False,
parallel=False,
partition_by=None,
extension=None,
key_id=None
) %}

{{ redshift.unload_table(
schema,
table,
s3_path,
iam_role=iam_role,
aws_key=aws_key,
aws_secret=aws_secret,
aws_region=aws_region,
aws_token=aws_token,
manifest=manifest,
header=header,
format=format,
delimiter=delimiter,
null_as=null_as,
max_file_size=max_file_size,
escape=escape,
compression=compression,
add_quotes=add_quotes,
encrypted=encrypted,
overwrite=overwrite,
cleanpath=cleanpath,
parallel=parallel,
partition_by=partition_by,
extension=extension
) }}

{% if key_id %}
KMS_KEY_ID '{{ key_id }}'
{% endif %}
{% endmacro %}