feat: replace callgraph dump with streaming CSV trace output

Replace callgrind's accumulated callgraph output with streaming CSV
trace data emitted at function ENTER/EXIT boundaries. Each row contains
delta counters since the last sample, enabling per-call cost attribution.

Key changes:
- dump.c: Replace callgraph output with CSV trace (trace_open/emit/close)
- callstack.c: Hook push/pop_call_stack to emit ENTER/EXIT samples
- threads.c: Add per-thread last_sample_cost for delta tracking
- global.h: Add trace_output struct and per-thread sample state
- main.c: Open trace at init, close at fini, update copyright

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

Author: art049

.gitignore

@@ -10,13 +10,16 @@
 /autom4te.cache
 /bin
 /cachegrind.out.*
+/callgrind.out.*
+/tracegrind.out.*
 /compile
 /config.guess
 /config.h*
 /config.log
 /config.status
 /config.sub
 /configure
+/configure~
 /default.supp
 /depcomp
 /glibc-2.X.supp
@@ -161,6 +164,30 @@
 /callgrind/tests/inline-samefile
 /callgrind/tests/inline-crossfile
 
+# /tracegrind/
+/tracegrind/*.so
+/tracegrind/.deps
+/tracegrind/tracegrind-*-darwin
+/tracegrind/tracegrind-*-linux
+/tracegrind/tracegrind-*-solaris
+/tracegrind/tracegrind-*-freebsd
+/tracegrind/Makefile
+/tracegrind/Makefile.in
+
+# /tracegrind/tests/
+/tracegrind/tests/*.dSYM
+/tracegrind/tests/*.post.diff*
+/tracegrind/tests/*.post.out
+/tracegrind/tests/*.stderr.diff*
+/tracegrind/tests/*.stderr.out
+/tracegrind/tests/*.stdout.diff*
+/tracegrind/tests/*.stdout.out
+/tracegrind/tests/.deps
+/tracegrind/tests/Makefile
+/tracegrind/tests/Makefile.in
+/tracegrind/tests/tracegrind.out.*
+/tracegrind/tests/fibo
+
 # /coregrind/
 /coregrind/*.a
 /coregrind/*.dSYM

tracegrind/callstack.c

@@ -250,6 +250,12 @@ void TG_(push_call_stack)(BBCC* from, UInt jmp, BBCC* to, Addr sp, Bool skip)
 
     TG_(current_call_stack).sp++;
 
+    /* Emit CSV trace sample on function entry */
+    if (!skip && TG_(current_state).collect) {
+	fn_node* to_fn = to->cxt->fn[0];
+	TG_(trace_emit_sample)(TG_(current_tid), "ENTER", to_fn);
+    }
+
     /* To allow for above assertion we set context of next frame to 0 */
     TG_ASSERT(TG_(current_call_stack).sp < TG_(current_call_stack).size);
     current_entry++;
@@ -353,6 +359,11 @@ void TG_(pop_call_stack)(void)
 	}
 	TG_(stat).ret_counter++;
 
+	/* Emit CSV trace sample on function exit */
+	if (TG_(current_state).collect) {
+	    TG_(trace_emit_sample)(TG_(current_tid), "EXIT", to_fn);
+	}
+
 	/* restore context */
 	TG_(current_state).cxt  = lower_entry->cxt;
 	TG_(current_fn_stack).top =