fix: recover stampissuer on batch creation

[aloknerurkar]

Currently, we create a new stampissuer when the user buys new stamps using the /stamps API.

This makes a call to the blockchain to buy a new postage batch. The operation can take some time and if for some reason the node goes down at this point, we might end up in a state where we have bought the postage batch, but a new stampissuer was not created making this batch not useful.

The change ensures that on batch creation event handled by the listener when we get the update from the blockchain, we will check if a stampissuer is present. And if it's not, we will create it.

---

This change is 

[acud]

Great stuff! thank you!

[zelig]

Reviewed 28 of 28 files at r1.
Reviewable status: 6 of 28 files reviewed, 2 unresolved discussions (waiting on @acud, @aloknerurkar, @esadakar, and @zelig)

---

pkg/postage/service.go, line 92 at r1 (raw file):

	case errors.Is(err, ErrNotFound):
		ps.Add(NewStampIssuer(
			"recovered",

what is this supposed to be?

---

pkg/postage/service_test.go, line 117 at r1 (raw file):

			t.Fatalf("expected no error, got %v", err)
		}
		if st.Label() != "recovered" {

label should not be used here

[acud]

Reviewed 6 of 28 files at r1, 22 of 22 files at r2.
Reviewable status: all files reviewed, 2 unresolved discussions (waiting on @esadakar and @zelig)

---

pkg/postage/service.go, line 92 at r1 (raw file):

Previously, zelig (Viktor Trón) wrote…

what is this supposed to be?

an indication that the batch was recovered by the listener

---

pkg/postage/service_test.go, line 117 at r1 (raw file):

Previously, zelig (Viktor Trón) wrote…

label should not be used here

@zelig I think that using the label here is just fine. Not sure what's the problem of using this field to indicate that the batch was recovered by the listener

[zelig]

Reviewed 3 of 22 files at r2.
Reviewable status: all files reviewed, 1 unresolved discussion (waiting on @esadakar and @zelig)