Body match-and-replace rules

Extends match-and-replace rules to request/response bodies, not just
headers. A body rule materializes the body into memory (bounded by
the same maxCaptureBytes cap as history capture) instead of streaming
it straight through - the opposite of the normal path, so it's only
paid when a body rule is actually configured. A body over the cap
passes through byte-exact and unmodified rather than being partially
rewritten.

Response Content-Length is recomputed explicitly when a rule changes
body length: unlike http.Request.Write, http.ResponseWriter doesn't
derive it from resp.ContentLength on its own, so a stale header would
otherwise corrupt response framing for the client.

The history audit trail still shows the original, pre-rule bytes on
both legs; only the wire traffic reflects the rewrite. Verified live
against a real daemon: origin receives the rewritten request body,
client receives the rewritten response body with correct
Content-Length, and history keeps the unmodified bytes.

Adds a Part selector (header/body) to the Rules add/edit form and
table in the TUI.

1 files changed, 39 insertions, 4 deletions

diff --git a/internal/rules/rules.go b/internal/rules/rules.go
index 1f50ec4..e0547db 100644
--- a/internal/rules/rules.go
+++ b/internal/rules/rules.go
@@ -1,7 +1,6 @@
 // Package rules implements match-and-replace: user-defined rules that
-// rewrite request/response headers as they pass through the proxy.
-// Deliberately headers-only for now - see ApplyHeaders for why bodies
-// are a separate, harder problem (noted as a follow-up in PLAN.md).
+// rewrite request/response headers or bodies as they pass through the
+// proxy. See ApplyHeaders and ApplyBody for how each part is handled.
 package rules
 
 import (
@@ -19,7 +18,7 @@ type Rule struct {
 	Enabled bool
 	Name    string
 	Scope   string // "request" or "response"
-	Part    string // "header" (only part supported so far)
+	Part    string // "header" or "body"
 	Match   string
 	Replace string
 	IsRegex bool
@@ -75,6 +74,42 @@ func ApplyHeaders(h http.Header, rs []Rule) http.Header {
 	return http.Header(mh)
 }
 
+// HasBodyRules reports whether any of rs (already filtered to enabled
+// rules for one scope, as returned by store.EnabledRules) is a body
+// rule - used to decide whether materializing a request/response body
+// at all is worth it. Most traffic has no body rule configured and
+// should keep streaming through unbuffered, which is what makes exact
+// capture of arbitrarily large bodies possible in the first place.
+func HasBodyRules(rs []Rule) bool {
+	for _, r := range rs {
+		if r.Enabled && r.Part == "body" {
+			return true
+		}
+	}
+	return false
+}
+
+// ApplyBody rewrites body by running every enabled rule with
+// Part=="body" over it, in Position order - literal substring replace
+// by default, or regex if the rule's Regex toggle is on. Unlike
+// ApplyHeaders, this operates directly on raw bytes with no reparse/
+// validation step: body content isn't a fixed format the way a header
+// block is, so there's nothing to parse back and confirm is still
+// well-formed. A rule that doesn't match anything is simply a no-op,
+// same as ApplyHeaders.
+func ApplyBody(body []byte, rs []Rule) []byte {
+	text := string(body)
+	for _, r := range sortedByPosition(rs) {
+		if !r.Enabled || r.Part != "body" {
+			continue
+		}
+		if next, ok := apply(text, r); ok {
+			text = next
+		}
+	}
+	return []byte(text)
+}
+
 func sortedByPosition(rs []Rule) []Rule {
 	out := make([]Rule, len(rs))
 	copy(out, rs)