Optimize C2 VerifyIterativeGVN execution time

When fixing IGVN related bug JDK-8265767, try verification with -XX:+VerifyIterativeGVN and it is extremly slow ("-Xbatch -XX:-TieredCompilation -XX:+VerifyIterativeGVN"). Checking VerifyIterativeGVN implementation, there exists spaces to optimize its execution time and make running -XX:+VerifyIterativeGVN possible when changing IGVN related code.

1. Optimize redundant verfications in PhaseIterGVN::verify_step. Nodes might verified multiple times.
    Redundant verifications between full pass and _verify_window single node process.
    Redundant verifications between different nodes in _verify_window

2. Optimize def-use edge checking
    Skip multiple check for same input node x
    Skip redundant check for index "i" in inner loop.

2238 for (uint i = 0; i < n->len(); i++) {
2243
2244 // Verify my input has a def-use edge to me
2245 // Count use-def edges from n to x
2246 int cnt = 0;
2247 for (uint j = 0; j < n->len(); j++) {
2248 if (n->in(j) == x) {
2249 cnt++;
2250 }
2251 }
2252
2253 // Count def-use edges from x to n
2254 uint max = x->_outcnt;
2255 for (uint k = 0; k < max; k++) {
2256 if (x->_out[k] == n) {
2257 cnt--;
2258 }
2259 }

3. Optimize implementation
    Replace "n->in(j)" with "n->_in[j]", similar with x->n processing, skip unuseful assert.
    

2247 for (uint j = 0; j < n->len(); j++) {
2248 if (n->in(j) == x) {
2249 cnt++;
2250 }
2251 }
2252
2253 // Count def-use edges from x to n
2254 uint max = x->_outcnt;
2255 for (uint k = 0; k < max; k++) {
2256 if (x->_out[k] == n) {
2257 cnt--;
2258 }
2259 }