New MS MARCO (V1) passage regressions

[lintool]

Just for consistency re: #1721 - we are going to re-build MS MARCO passage regressions, using same scripts, for consistency.

Raw file that @ronakice has prepped on orca is here:

$ md5sum /store/scratch/rpradeep/msmarco-v1/collections/msmarco_v1_passage_d2q-t5_40/docs.jsonl
1f4859d1258df07b8ac2277a986e162b  /store/scratch/rpradeep/msmarco-v1/collections/msmarco_v1_passage_d2q-t5_40/docs.jsonl

This is the starting point. I'll repackage in the same way as doc regressions.

[ronakice]

Yep, looks good from my end, @MXueguang should check if it looks fine on his!

[lintool]

Repackaged and stored on orca at /store/collections/msmarco/tarballs:

$ md5sum msmarco-passage-docTTTTTquery.tar
ee1a1d9e89d26bae4c0382d75a361576  msmarco-passage-docTTTTTquery.tar

Unpacked at /store/collections/msmarco/msmarco-passage-docTTTTTquery:

$ ls -l
total 3156412
-r--r--r-- 1 jimmylin jimmylin 180510918 Jan 13 13:12 docs-aa.jsonl.gz
-r--r--r-- 1 jimmylin jimmylin 180464659 Jan 13 13:12 docs-ab.jsonl.gz
-r--r--r-- 1 jimmylin jimmylin 180596698 Jan 13 13:12 docs-ac.jsonl.gz
-r--r--r-- 1 jimmylin jimmylin 180995768 Jan 13 13:12 docs-ad.jsonl.gz
-r--r--r-- 1 jimmylin jimmylin 181972324 Jan 13 13:12 docs-ae.jsonl.gz
-r--r--r-- 1 jimmylin jimmylin 183045317 Jan 13 13:12 docs-af.jsonl.gz
-r--r--r-- 1 jimmylin jimmylin 183262993 Jan 13 13:12 docs-ag.jsonl.gz
-r--r--r-- 1 jimmylin jimmylin 183505225 Jan 13 13:12 docs-ah.jsonl.gz
-r--r--r-- 1 jimmylin jimmylin 183467390 Jan 13 13:12 docs-ai.jsonl.gz
-r--r--r-- 1 jimmylin jimmylin 183563636 Jan 13 13:12 docs-aj.jsonl.gz
-r--r--r-- 1 jimmylin jimmylin 183872883 Jan 13 13:12 docs-ak.jsonl.gz
-r--r--r-- 1 jimmylin jimmylin 184319850 Jan 13 13:12 docs-al.jsonl.gz
-r--r--r-- 1 jimmylin jimmylin 182176900 Jan 13 13:12 docs-am.jsonl.gz
-r--r--r-- 1 jimmylin jimmylin 183767988 Jan 13 13:12 docs-an.jsonl.gz
-r--r--r-- 1 jimmylin jimmylin 183210261 Jan 13 13:12 docs-ao.jsonl.gz
-r--r--r-- 1 jimmylin jimmylin 184168860 Jan 13 13:13 docs-ap.jsonl.gz
-r--r--r-- 1 jimmylin jimmylin 183566373 Jan 13 13:13 docs-aq.jsonl.gz
-r--r--r-- 1 jimmylin jimmylin 125664655 Jan 13 13:13 docs-ar.jsonl.gz

[lintool]

Confirmed on orca, passes without issue with :

$ python src/main/python/run_regression.py --index --verify --search --regression msmarco-passage-docTTTTTquery