Fix #1204: Correct #nosec handling for f-strings and multi-line statements

[gouriphadnis0301]

Description:
This pull request addresses Issue #1204 in Bandit, where # nosec BXXX comments were not correctly recognized on f-strings or multi-line statements.

Problem:

• The existing get_nosec() function in utils.py only checked line numbers (context["linerange"]) to determine if a # nosec comment was present.

• This approach caused false positives: Bandit incorrectly raised warnings for code lines that had # nosec when the comment was part of a multi-line f-string or expression.

• Specific rule IDs like B607, B602 in # nosec BXXX comments were not properly parsed.

Fix:

1. Replaced the old get_nosec() function with a new implementation that:
2. Iterates over all lines in context["linerange"] to detect # nosec.
3. Properly handles multi-line statements and f-strings.
4. Parses specific rule IDs (e.g., B607, B602) and returns them as a list.
5. Returns True for a general # nosec without rule IDs.
6. Returns None if no # nosec comment is found.

Impact:

• Bandit will now ignore the correct lines with # nosec for both f-strings and multi-line expressions.
• Reduces false positive security warnings and improves accuracy.

Testing:

• Verified that lines with # nosec and rule IDs (# nosec B607) are now correctly ignored.
• Checked multi-line f-strings and concatenated expressions — Bandit no longer reports unnecessary warnings.
• Ran existing Bandit tests:
  pytest -k nosec
• All tests pass successfully.