fix: remove network-related packages from preimport warmup

chardet, charset_normalizer, requests, and httpx crash the guest
with SIGSEGV during pyhl setup warmup (likely due to urllib3/ssl/anyio
init). Keep them in the rootfs as shipped-but-not-preimported.
Retain python-docx and python-pptx in the preimport list.

Signed-off-by: danbugs <danilochiarlone@gmail.com>

Author: danbugs

docs/python-packages.md

@@ -38,11 +38,8 @@ These packages are imported during `pyhl setup` warmup. They are already in `sys
 | Package | Import name |
 |---------|-------------|
 | beautifulsoup4 | `bs4` |
-| chardet | `chardet` |
-| charset-normalizer | `charset_normalizer` |
 | click | `click` |
 | cryptography | `cryptography` |
-| httpx | `httpx` |
 | Jinja2 | `jinja2` |
 | lxml | `lxml` |
 | markdown-it-py | `markdown_it` |
@@ -57,7 +54,6 @@ These packages are imported during `pyhl setup` warmup. They are already in `sys
 | python-dotenv | `dotenv` |
 | python-pptx | `pptx` |
 | PyYAML | `yaml` |
-| requests | `requests` |
 | tabulate | `tabulate` |
 | tenacity | `tenacity` |
 | tqdm | `tqdm` |
@@ -76,6 +72,8 @@ These packages are in the rootfs but not pre-imported. The first `import` pays t
 | boto3 | `boto3` |
 | builtwith | `builtwith` |
 | celery | `celery` |
+| chardet | `chardet` |
+| charset-normalizer | `charset_normalizer` |
 | coverage | `coverage` |
 | distro | `distro` |
 | docx2txt | `docx2txt` |
@@ -91,6 +89,7 @@ These packages are in the rootfs but not pre-imported. The first `import` pays t
 | gitpython | `git` |
 | google-api-python-client | `googleapiclient` |
 | hypercorn | `hypercorn` |
+| httpx | `httpx` |
 | hypothesis | `hypothesis` |
 | loguru | `loguru` |
 | markdown | `markdown` |
@@ -124,6 +123,7 @@ These packages are in the rootfs but not pre-imported. The first `import` pays t
 | rapidfuzz | `rapidfuzz` |
 | rarfile | `rarfile` |
 | reportlab | `reportlab` |
+| requests | `requests` |
 | rope | `rope` |
 | ruff | `ruff` |
 | schedule | `schedule` |

examples/python-agent-driver/hl_pydriver.c

@@ -253,8 +253,7 @@ static void py_initialize_once(void)
 		"    'bs4', 'tabulate', 'click', 'tenacity', 'tqdm',"
 		"    'openpyxl', 'pypdf', 'markdown_it', 'PIL', 'lxml',"
 		"    'cryptography', 'dateutil', 'dotenv',"
-		"    'docx', 'pptx', 'chardet', 'charset_normalizer',"
-		"    'requests', 'httpx'):\n"
+		"    'docx', 'pptx'):\n"
 		"  try:\n"
 		"    importlib.import_module(_mod)\n"
 		"  except Exception as _e:\n"

host/src/bin/pyhl.rs

@@ -91,10 +91,6 @@ const PREIMPORTED_MODULES: &[&str] = &[
     "dotenv",
     "docx",
     "pptx",
-    "chardet",
-    "charset_normalizer",
-    "requests",
-    "httpx",
 ];
 
 /// Build the long-about blurb shown by `pyhl --help`. Lists the
@@ -122,19 +118,20 @@ fn long_about() -> String {
         "\n\n\
          Additional packages shipped in the rootfs (pay import cost on \
          first use): aiohttp, altair, APScheduler, bandit, bokeh, \
-         boto3, builtwith, celery, coverage, distro, docx2txt, duckdb, \
-         edgartools, exchange-calendars, fabric, Faker, fastapi, \
-         feedparser, fpdf2, gensim, gitpython, \
-         google-api-python-client, hypercorn, hypothesis, loguru, \
-         markdown, markdownify, mutagen, networkx, nltk, \
-         numpy-financial, odfpy, paramiko, pdfplumber, pdfrw, pexpect, \
-         pipdeptree, platformdirs, plotly, polars, praw, pycountry, \
-         pydub, pyflakes, pygments, pylint, PyPDF2, pytest, \
-         pytest-asyncio, pytest-cov, pyxlsb, qrcode, radon, rapidfuzz, \
-         rarfile, reportlab, rope, ruff, schedule, scikit-learn, scipy, \
-         scrapy, send2trash, slack-sdk, srt, statsmodels, svgwrite, \
-         sympy, textblob, trafilatura, tweepy, typer, uvicorn, vulture, \
-         watchdog, websockets, wordcloud, xlrd, xlsxwriter.\n\n\
+         boto3, builtwith, celery, chardet, charset-normalizer, \
+         coverage, distro, docx2txt, duckdb, edgartools, \
+         exchange-calendars, fabric, Faker, fastapi, feedparser, \
+         fpdf2, gensim, gitpython, google-api-python-client, \
+         hypercorn, httpx, hypothesis, loguru, markdown, markdownify, \
+         mutagen, networkx, nltk, numpy-financial, odfpy, paramiko, \
+         pdfplumber, pdfrw, pexpect, pipdeptree, platformdirs, plotly, \
+         polars, praw, pycountry, pydub, pyflakes, pygments, pylint, \
+         PyPDF2, pytest, pytest-asyncio, pytest-cov, pyxlsb, qrcode, \
+         radon, rapidfuzz, rarfile, reportlab, requests, rope, ruff, \
+         schedule, scikit-learn, scipy, scrapy, send2trash, slack-sdk, \
+         srt, statsmodels, svgwrite, sympy, textblob, trafilatura, \
+         tweepy, typer, uvicorn, vulture, watchdog, websockets, \
+         wordcloud, xlrd, xlsxwriter.\n\n\
          Packages not in the rootfs will raise ModuleNotFoundError.",
     );
     s