Refactor sorted map encode to use fewer buffers for nested maps.

[benluddy]

Description

Runs a bit faster, but more importantly, only needs a single buffer to encode nested, sorted maps
instead of using multiple temporary buffers.

                                                                │ before.txt  │             after.txt              │
                                                                │   sec/op    │   sec/op     vs base               │
    MarshalCanonical/Go_map[string]string_to_CBOR_map_canonical   1.464µ ± 0%   1.395µ ± 0%  -4.68% (p=0.000 n=10)
    MarshalCanonical/Go_map[int]int_to_CBOR_map_canonical         192.1n ± 0%   186.2n ± 1%  -3.10% (p=0.000 n=10)
    geomean                                                       530.2n        509.6n       -3.89%
    
                                                                │ before.txt │               after.txt               │
                                                                │    B/op    │    B/op      vs base                  │
    MarshalCanonical/Go_map[string]string_to_CBOR_map_canonical   88.00 ± 0%   112.00 ± 0%  +27.27% (p=0.000 n=10)
    MarshalCanonical/Go_map[int]int_to_CBOR_map_canonical         3.000 ± 0%    3.000 ± 0%        ~ (p=1.000 n=10) ¹
    geomean                                                       16.25         18.33       +12.82%
    ¹ all samples are equal
    
                                                                │ before.txt │              after.txt              │
                                                                │ allocs/op  │ allocs/op   vs base                 │
    MarshalCanonical/Go_map[string]string_to_CBOR_map_canonical   2.000 ± 0%   2.000 ± 0%       ~ (p=1.000 n=10) ¹
    MarshalCanonical/Go_map[int]int_to_CBOR_map_canonical         1.000 ± 0%   1.000 ± 0%       ~ (p=1.000 n=10) ¹
    geomean                                                       1.414        1.414       +0.00%
    ¹ all samples are equal

PR Was Proposed and Welcomed in Currently Open Issue

• This PR was proposed and welcomed by maintainer(s) in issue #___
• Closes or Updates Issue #___

Checklist (for code PR only, ignore for docs PR)

• Include unit tests that cover the new code
• Pass all unit tests
• Pass all lint checks in CI (goimports, gosec, staticcheck, etc.)
• Sign each commit with your real name and email.
  Last line of each commit message should be in this format:
  Signed-off-by: Firstname Lastname firstname.lastname@example.com
• Certify the Developer's Certificate of Origin 1.1
  (see next section).

Certify the Developer's Certificate of Origin 1.1

• By marking this item as completed, I certify
  the Developer Certificate of Origin 1.1.

Developer Certificate of Origin
Version 1.1

Copyright (C) 2004, 2006 The Linux Foundation and its contributors.
660 York Street, Suite 102,
San Francisco, CA 94110 USA

Everyone is permitted to copy and distribute verbatim copies of this
license document, but changing it is not allowed.

Developer's Certificate of Origin 1.1

By making a contribution to this project, I certify that:

(a) The contribution was created in whole or in part by me and I
    have the right to submit it under the open source license
    indicated in the file; or

(b) The contribution is based upon previous work that, to the best
    of my knowledge, is covered under an appropriate open source
    license and I have the right under that license to submit that
    work with modifications, whether created in whole or in part
    by me, under the same open source license (unless I am
    permitted to submit under a different license), as indicated
    in the file; or

(c) The contribution was provided directly to me by some other
    person who certified (a), (b) or (c) and I have not modified
    it.

(d) I understand and agree that this project and the contribution
    are public and that a record of the contribution (including all
    personal information I submit with it, including my sign-off) is
    maintained indefinitely and may be redistributed consistent with
    this project or the open source license(s) involved.

[fxamacker]

Thanks @benluddy for opening this PR!

It looks good and I left a few comments that Less() used for sorting map elements should use valueOffset instead of nextOffset to sort map keys.

[fxamacker]

We should only compare keys (not entire map element).

Suggested change

	return bytes.Compare(x.data[kvi.offset:kvi.nextOffset], x.data[kvj.offset:kvj.nextOffset]) <= 0
	return bytes.Compare(x.data[kvi.offset:kvi.valueOffset], x.data[kvj.offset:kvj.valueOffset]) <= 0

[benluddy]

Great catch, thanks! Fixed, and tweaked the test covering this slightly so that it will fail if the length of the entire entry is compared instead of the length of the key only.

[fxamacker]

LGTM! Thanks for super fast response and updating this PR on a Sunday! 👍