Fix #1385: derive nxcdb hosts CSV header from the protocol's HostsTable

[Diogo-Damasceno]

Description

Fixes #1385. The export hosts CSV header was hard-coded to the SMB schema (smbv1/signing/spooler/zerologon/petitpotam), so exporting hosts for any other protocol (RDP, MSSQL, SSH, LDAP, ...) produced a header that did not match the row content.

Change: the header is now derived from the active protocol' HostsTable columns via a new host_csv_headers() helper, as suggested in the issue (default to the real column names instead of static SMB strings). simple keeps the first 8 columns (historical row width); detailed returns every column of the table. I kept the diff to just this logic — no autoformatter noise on unrelated lines.

AI usage disclosure: I put this together with the help of the DeepHat agent and the Hermes Agent, which I used strictly as tools. I drove every step myself: identified the root cause, reviewed the diff line by line, wrote and ran the regression tests, and confirmed the SMB simple/detailed output shape is unchanged (no regression). The logic and the tests are mine; the agents were just a faster keyboard.

Type of change

• Bug fix (non-breaking change which fixes an issue)

Setup guide for the review

• Build a workspace with, e.g., RDP hosts: nxc rdp <host> then nxcdb, proto rdp, export hosts detailed rdp.csv.
  • Before: header was id;ip;hostname;domain;os;dc;smbv1;signing;spooler;zerologon;petitpotam (SMB-only, wrong for RDP).
  • After: header is id;ip;hostname;domain;os;nla -- matches the actual RDP row content.
• SMB export is unchanged in shape (simple still = first 8 SMB columns; detailed = all 11).
• Local test: pytest tests/test_nxcdb_host_export.py (3 tests: SMB simple/detailed unchanged, RDP has no SMB-only columns). Note: I could not run the full NetExec suite locally because aardwolf does not build on Python 3.14 (see Install fails with Python 3.14 #1241); the change is isolated to the CSV header derivation and is covered by the new unit tests.

Checklist

• I have ran Ruff against my changes
• I have added or updated the tests (tests/test_nxcdb_host_export.py)
• I have linked relevant sources (issue nxcdb CSV header mismatch #1385; maintainer direction: use the real column names)
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• This requires a documentation update (no user-facing behaviour change beyond the corrected header)

[NeffIsBack]

Hi and thanks for the PR!

Please revert the formatting changes tho. Also please disclose AI usage in the PR description (see PR template).

[Diogo-Damasceno]

Thanks @NeffIsBack — and sorry about the formatting noise, that was on me. I let the formatter loose on lines that had nothing to do with the fix.

I've reverted all of it now: the diff is just the host_csv_headers() helper and its use in do_export. No unrelated reformatting.

I also added the AI-usage disclosure to the description as you asked. For the record: I put this together with the help of the DeepHat agent and the Hermes Agent, which I used strictly as tools — I drove every step myself, reviewed the diff line by line, wrote and ran the regression tests, and confirmed the SMB simple/detailed shape is unchanged.

Logic and tests are unchanged; SMB exports still produce the same simple/detailed shape as before. Let me know if anything else needs a look.