fix: pass re.DOTALL as keyword arg in Livy comment-stripping regex

[sdebruyn]

Summary

Fixes #194.

_getLivySQL (in singleton_livy.py) and _strip_block_comments (in concurrent_livy.py) both call:

re.sub(r"\s*/\*(.|\n)*?\*/\s*", "\n", sql, re.DOTALL)

re.DOTALL is the integer enum value 16, and re.sub's positional signature is re.sub(pattern, repl, string, count=0, flags=0). Passing re.DOTALL positionally sets count=16, capping comment-stripping to at most 16 replacements per submitted SQL string.

This PR swaps both call sites to flags=re.DOTALL so the comment-stripping pass is no longer capped and the flag the author intended to set is actually set.

Diff

-re.sub(r"\s*/\*(.|\n)*?\*/\s*", "\n", sql, re.DOTALL)
+re.sub(r"\s*/\*(.|\n)*?\*/\s*", "\n", sql, flags=re.DOTALL)

Applied identically to both singleton_livy.py:488 and concurrent_livy.py:555.

Test plan

• CI green on this branch
• Manual: submit a model with >16 /* ... */ comment blocks via Livy; verify all comments are stripped before submission (was: 4 surviving comments with the original count=16 cap; now: 0).

Notes

A natural follow-up would be to extract a single _strip_sql_comments(sql) helper used by both files so this duplicated regex (and its history of being miswritten) can't drift again. Left out of this PR to keep the diff minimal and easy to backport.

[sdebruyn]

@microsoft-github-policy-service agree

[mdrakiburrahman]

Same as here:

#193 (comment)

[mdrakiburrahman]

I think this one is simple enough where I'll run CI locally and merge if it runs green 🙂

[mdrakiburrahman]

I have fired CI on fs-04:

Will share results here after running.

[mdrakiburrahman]

CI ran green on the final commit in this branch: