Support parsing pgwire timestamps as they are sent by lib/pq

[danhhz]

Closes #5612.

This change is

[danhhz]

After this lands, I'm going to revert cockroachdb/pq@a7769c7 as it was the wrong fix and is unnecessary (sorry, I should have gotten the entire context before committing anything).

[dt]

Review status: 0 of 2 files reviewed at latest revision, 1 unresolved discussion.

---

sql/pgwire/types.go, line 255 [r1] (raw file):
Is this exercised by any tests currently?

---

Comments from Reviewable

[danhhz]

Review status: 0 of 2 files reviewed at latest revision, 1 unresolved discussion.

---

sql/pgwire/types.go, line 255 [r1] (raw file):
Nope, could be dead code for all I know. I was just keeping it because it allows roundtripping the timestamp bytes that CockroachDB sends back. I could add a test for the round trip though.

---

Comments from Reviewable

[petermattis]

Review status: 0 of 2 files reviewed at latest revision, 2 unresolved discussions.

---

sql/pgwire/types_test.go, line 26 [r1] (raw file):
You can define this variable inside of TestParseTs.

---

Comments from Reviewable

[petermattis]

Review status: 0 of 2 files reviewed at latest revision, 3 unresolved discussions.

---

sql/pgwire/types_test.go, line 31 [r1] (raw file):
Add a test for the format accepted by pq.ParseTimestamp?

---

Comments from Reviewable

[petermattis]

---

Review status: 0 of 2 files reviewed at latest revision, 3 unresolved discussions.

---

Comments from Reviewable

[danhhz]

Review status: 0 of 2 files reviewed at latest revision, 3 unresolved discussions, all commit checks successful.

---

sql/pgwire/types_test.go, line 26 [r1] (raw file):
Done.

---

sql/pgwire/types_test.go, line 31 [r1] (raw file):
Done.

---

Comments from Reviewable