feat(suggest): add Python dependency parser for deep framework detection

[yacosta738]

Context

The current detection system is heavily oriented around package.json for dependency scanning. Python projects use pyproject.toml, requirements.txt, Pipfile, or setup.py — none of which are parsed for dependency names today. This blocks proper detection of Django, Flask, FastAPI, Pytest, LangChain, OpenAI, PyTorch, TensorFlow, and many more.

Proposal

Add Python dependency parsing analogous to collect_package_names for package.json:

New parsers needed in detect.rs:

1. parse_requirements_txt — Line-by-line, strip version specifiers (django>=4.0 → django)
2. parse_pyproject_toml_deps — Parse [project.dependencies] and [project.optional-dependencies] from pyproject.toml
3. parse_pipfile_deps (optional) — Parse [packages] and [dev-packages] from Pipfile

New detection field:

Either add a python_packages detection method or reuse config_file_content with smarter patterns.

Technologies This Unlocks

Technology	Python Package
Django	django
Flask	flask
FastAPI	fastapi
Pytest	pytest
LangChain	langchain
OpenAI	openai
PyTorch	torch
TensorFlow	tensorflow

Acceptance Criteria

• requirements.txt dependencies are parsed and matched
• pyproject.toml dependencies are parsed and matched
• Django/Flask/FastAPI detected in respective projects
• Unit tests for each parser
• cargo test suggest passes

[linear]

DALLAY-239