Fix route coordinate order for CMU tasks

[ctkrug]

Summary\n\n- correct OSRM coordinate order for routing tasks 737–741 so Carnegie Mellon University is the origin\n- add a regression test covering all five tasks\n\n## Validation\n\n- ============================= test session starts ==============================

platform darwin -- Python 3.11.15, pytest-9.0.1, pluggy-1.6.0
rootdir: /Users/Krug/Desktop/ChatGPT/webarena-verified
configfile: pyproject.toml
plugins: playwright-0.7.2, anyio-4.12.0, base-url-2.1.0, rerunfailures-16.1, cov-7.0.0
collected 1820 items

tests/dataset/test_dataset_integrity.py ................................ [ 1%]
........................................................................ [ 5%]
........................................................................ [ 9%]
........................................................................ [ 13%]
........................................................................ [ 17%]
........................................................................ [ 21%]
........................................................................ [ 25%]
........................................................................ [ 29%]
........................................................................ [ 33%]
........................................................................ [ 37%]
........................................................................ [ 41%]
........................................................................ [ 45%]
........................................................................ [ 49%]
........................................................................ [ 53%]
........................................................................ [ 57%]
........................................................................ [ 61%]
........................................................................ [ 65%]
........................................................................ [ 69%]
........................................................................ [ 72%]
........................................................................ [ 76%]
........................................................................ [ 80%]
........................................................................ [ 84%]
........................................................................ [ 88%]
........................................................................ [ 92%]
........................................................................ [ 96%]
............................................................ [100%]

============================= 1820 passed in 1.54s ============================= (1,820 passed)\n-
╭──────────────────────────────────── LINT ────────────────────────────────────╮
│ │
╰──────────────────────────────────────────────────────────────────────────────╯

Running ruff check...
All checks passed!
Running ruff format check...
195 files already formatted
Running ty check...
All checks passed!
Running actionlint...

✓ All checks passed\n- ruff (legacy alias)......................................................Passed
ruff format..............................................................Passed
ty.......................................................................Passed\n\nCloses #6