Fix: Prevent false positives in expertise keyword matching

[Atharva-1512]

Replaced substring matching with regex word boundary matching to avoid incorrect matches (e.g., "law" matching "flaw").

• Uses re.search with word boundaries
• Case-insensitive matching
• Improves expertise scoring accuracy

Fixes #4

[rosspeili]

Hi @Atharva-1512 ! Thanks so much for taking the time to submit this PR and for looking into Issue #4.

After reviewing the specific changes here, we won't be able to merge this one for a few reasons:

1. The Issue is already fixed in main: If you check the current codebase (rooms/session.py), the strict \b word boundary logic has actually already been implemented in a previous commit!
   if re.search(rf"\b{re.escape(kw.lower())}\b", context_lower):
2. Redundant Regex Logic: The proposed change adds re.IGNORECASE to the regex search. However, because the text is specifically lowercased right above it (context_lower = context_text.lower()), the IGNORECASE flag is redundant against the already-lowercased string and won't improve performance.
3. .gitignore Additions: This PR bundles in a change to .gitignore for .history/. We prefer to keep personal IDE-specific folders (like VS Code's Local History) out of the project's global .gitignore to keep commits focused strictly on the feature or fix at hand.

Because the underlying bug has already been patched in main, I'm going to go ahead and close this PR. But thank you again for being willing to contribute, and feel free to check out our other open issues!

Thanks again. <3