issue #20 uint vs int for tree indices and sizes

Is there a reason the CBOR uses int (cose major type 1) in preference to
uint ? If not, I suggest we use uint to preserve the full range of
positive values possible.

note: I may need to update the dbor-diag output. If this change is
acceptable in principal I am happy to go do that as a requirement to
merge.

draft-ietf-cose-merkle-tree-proofs.md

@@ -277,10 +277,10 @@ The CBOR representation of an inclusion proof for RFC9162_SHA256 is:
 inclusion-proof = bstr .cbor [
 
     ; tree size at current merkle root
-    tree-size: int
+    tree-size: uint
 
     ; index of leaf in tree
-    leaf-index: int
+    leaf-index: uint
 
     ; path from leaf to current merkle root
     inclusion-path: [ + bstr ]
@@ -399,10 +399,10 @@ The cbor representation of a consistency proof for RFC9162_SHA256 is:
 consistency-proof =  bstr .cbor [
 
     ; previous merkle root tree size
-    tree-size-1: int
+    tree-size-1: uint
 
     ; latest merkle root tree size
-    tree-size-2: int
+    tree-size-2: uint
 
     ; path from previous merkle root to latest merkle root.
     consistency-path: [ + bstr ]