feat(13-01): implement NpmSource and PyPISource with httptest tests

- NpmSource searches npm registry JSON API for provider keywords
- PyPISource scrapes pypi.org search HTML for project links
- Both credentialless, rate-limited at 1 req/2s, burst 2
- httptest-based tests verify Sweep, ctx cancellation, Name/Rate/Burst

pkg/recon/sources/pypi.go

@@ -0,0 +1,102 @@
+package sources
+
+import (
+	"context"
+	"fmt"
+	"io"
+	"net/http"
+	"net/url"
+	"regexp"
+	"time"
+
+	"golang.org/x/time/rate"
+
+	"github.com/salvacybersec/keyhunter/pkg/providers"
+	"github.com/salvacybersec/keyhunter/pkg/recon"
+)
+
+// PyPISource searches pypi.org for packages matching provider keywords.
+// Scrapes the HTML search page since PyPI has no public search JSON API.
+// No credentials required. Emits findings tagged SourceType=recon:pypi.
+type PyPISource struct {
+	BaseURL  string
+	Registry *providers.Registry
+	Limiters *recon.LimiterRegistry
+	Client   *Client
+}
+
+var _ recon.ReconSource = (*PyPISource)(nil)
+
+// pypiProjectRE matches /project/{name}/ hrefs in search results.
+var pypiProjectRE = regexp.MustCompile(`^/project/[^/]+/?$`)
+
+func (s *PyPISource) Name() string            { return "pypi" }
+func (s *PyPISource) RateLimit() rate.Limit    { return rate.Every(2 * time.Second) }
+func (s *PyPISource) Burst() int               { return 2 }
+func (s *PyPISource) RespectsRobots() bool     { return false }
+func (s *PyPISource) Enabled(_ recon.Config) bool { return true }
+
+func (s *PyPISource) Sweep(ctx context.Context, _ string, out chan<- recon.Finding) error {
+	base := s.BaseURL
+	if base == "" {
+		base = "https://pypi.org"
+	}
+	client := s.Client
+	if client == nil {
+		client = NewClient()
+	}
+
+	queries := BuildQueries(s.Registry, "pypi")
+	if len(queries) == 0 {
+		return nil
+	}
+
+	for _, q := range queries {
+		if err := ctx.Err(); err != nil {
+			return err
+		}
+
+		if s.Limiters != nil {
+			if err := s.Limiters.Wait(ctx, s.Name(), s.RateLimit(), s.Burst(), false); err != nil {
+				return err
+			}
+		}
+
+		searchURL := fmt.Sprintf("%s/search/?q=%s", base, url.QueryEscape(q))
+		req, err := http.NewRequestWithContext(ctx, http.MethodGet, searchURL, nil)
+		if err != nil {
+			return fmt.Errorf("pypi: build req: %w", err)
+		}
+
+		resp, err := client.Do(ctx, req)
+		if err != nil {
+			return fmt.Errorf("pypi: fetch: %w", err)
+		}
+
+		hrefs, err := extractPyPIProjectLinks(resp.Body)
+		_ = resp.Body.Close()
+		if err != nil {
+			return fmt.Errorf("pypi: parse html: %w", err)
+		}
+
+		for _, href := range hrefs {
+			if err := ctx.Err(); err != nil {
+				return err
+			}
+			absURL := base + href
+			out <- recon.Finding{
+				ProviderName: "",
+				Source:       absURL,
+				SourceType:   "recon:pypi",
+				Confidence:   "low",
+				DetectedAt:   time.Now(),
+			}
+		}
+	}
+	return nil
+}
+
+// extractPyPIProjectLinks extracts unique /project/{name}/ hrefs from HTML.
+func extractPyPIProjectLinks(body io.Reader) ([]string, error) {
+	return extractAnchorHrefs(body, pypiProjectRE)
+}