Fix flytekit-greatexpectations tests: replace dead ge_tutorials CSV URL

[pingsutw]

Why are the changes needed?

build-plugins (flytekit-greatexpectations) is failing on master. The
superconductive/ge_tutorials GitHub repo no longer exists, so
https://raw.githubusercontent.com/superconductive/ge_tutorials/main/data/yellow_tripdata_sample_2019-01.csv
returns 404 and every test that feeds a remote FlyteFile/CSVFile into a
Great Expectations task blows up with:

FlyteDataNotFoundException: USER:ValueError: error=Value error!
Received: https://raw.githubusercontent.com/superconductive/ge_tutorials/main/data/yellow_tripdata_sample_2019-01.csv. File not found

6 failures: test_schema.py::test_ge_schema_remote_flytefile,
test_ge_schema_remote_flytefile_literal, test_task.py::test_ge_remote_flytefile,
test_ge_remote_flytefile_with_task, test_ge_remote_flytefile_workflow,
test_ge_flytefile_multiple_args.

What changes were proposed in this pull request?

Point those URLs at the copies of the same CSVs that are already committed in
this repo under plugins/flytekit-greatexpectations/tests/data/, served over
raw.githubusercontent.com from master. That keeps the tests exercising the
remote-http download path (which is the point of these tests) while removing
the dependency on a third-party repo we don't control.

Both files are byte-identical to the local ones the other tests already use
(md5 38f1366... for 2019-01, e39a6b8... for 2019-02), so the == 10000
row assertions are unchanged.

I also checked great-expectations/gx_tutorials as a replacement — it is live,
but its 2019-01 sample has 9999 rows instead of 10000 and its 2019-02 sample
differs, so it would have required changing the assertions.

How was this patch tested?

No new tests — this restores 6 existing tests that fail on master.

Verified the new URLs resolve and that flytekit's CSVFile download path
returns the expected row count:

$ curl -s -o /dev/null -w '%{http_code}' https://raw.githubusercontent.com/flyteorg/flytekit/master/plugins/flytekit-greatexpectations/tests/data/yellow_tripdata_sample_2019-01.csv
200

@task
def rows(dataset: CSVFile) -> int:
    return len(pd.read_csv(dataset))

@workflow
def wf(dataset: CSVFile = URL) -> int:
    return rows(dataset=dataset)

assert wf() == 10000   # passes

The full plugin suite needs great-expectations + pyspark 3.3.2, which I did not
install locally — CI on this PR is the real check.

Check all the applicable boxes

• I updated the documentation accordingly. (n/a — test-only change)
• All new and existing tests passed.
• All commits are signed-off.