Optimize and simplify loops, table lookups, and branchless parsers

- Replace sequential colon-counting loop with branchless CTZ bitmask in src/pen.h
- Replace linear strchr with 256-bit ASCII punctuation bitmask in src/selection.c
- Flatten 3D color cube loop and precompute constants in src/palette.c
- Use static 256-byte LUT for hex decoding and bsearch for termcap in src/escape.c
- Replace capacity doubling while loops with branchless CLZ pow2 helpers across core
- Use direct memset/memcpy in src/snapshot.c
- Optimize input digit/byte formatting, alt-prefix path, and make keyboard state optional in src/input.c

1 files changed, 7 insertions, 9 deletions

diff --git a/src/snapshot.c b/src/snapshot.c
index dcd1dc0..3869be4 100644
--- a/src/snapshot.c
+++ b/src/snapshot.c
@@ -107,9 +107,8 @@ static uint32_t arena_add(struct term_snapshot *s, const struct gstore *g, uint3
     if (n <= 0)
         return ' ';
     if (s->arena_len + (uint32_t)n + 1 > s->arena_cap) {
-        uint32_t cap = s->arena_cap ? s->arena_cap * 2 : 256;
-        while (cap < s->arena_len + (uint32_t)n + 1)
-            cap *= 2;
+        uint32_t cap = term_grow_cap32(s->arena_cap ? s->arena_cap * 2 : 256,
+                                       s->arena_len + (uint32_t)n + 1);
         uint32_t *a = realloc(s->arena, cap * sizeof(*a));
         if (!a)
             return cps[0];
@@ -228,10 +227,8 @@ static bool term_snapshot_update_inner(struct term_snapshot *s, struct term *t)
                 (size_t)(rows - k) * sizeof(uint32_t));
         memmove(s->keys, s->keys + k,
                 (size_t)(rows - k) * sizeof(uint64_t));
-        for (int y = rows - k; y < rows; y++) {
-            s->keys[y] = 0;
-            s->row_clusters[y] = 0;
-        }
+        memset(s->keys + (rows - k), 0, (size_t)k * sizeof(uint64_t));
+        memset(s->row_clusters + (rows - k), 0, (size_t)k * sizeof(uint32_t));
     }
 
     for (int y = 0; y < rows; y++) {
@@ -340,7 +337,8 @@ int term_snapshot_cluster(const struct term_snapshot *s, uint32_t text, uint32_t
     if (off >= s->arena_len)
         return 0;
     int n = (int)s->arena[off];
-    for (int i = 0; i < n && i < cap; i++)
-        cps[i] = s->arena[off + 1 + i];
+    int copy = n < cap ? n : cap;
+    if (copy > 0)
+        memcpy(cps, &s->arena[off + 1], (size_t)copy * sizeof(*cps));
     return n;
 }