NEWS.adoc: move ragtech entry to post-2.8.5 release

Signed-off-by: Jim Klimov <jimklimov+nut@gmail.com>

Author: jimklimov

NEWS.adoc

@@ -75,6 +75,35 @@ https://github.com/networkupstools/nut/milestone/13
       VP-1250-LCD) that return a 7-byte report instead of the standard 6 bytes,
       allowing the Megatec payload to be successfully reconstructed. [PR #3485]
 
+ - Introduced a new NUT driver named `ragtech` which provides support for the
+   Ragtech "Easy Pro" family of line-interactive UPS units (also sold under
+   the NEP, TORO, INNERGIE and OneUP brands in Brazil). Devices present a USB
+   CDC-ACM serial interface (VID `0x04D8`, PID `0x000A`) and speak a proprietary
+   6-byte register-access protocol with read/write/AND/OR opcodes; the driver
+   covers the 20 models in the family-10 device table and was validated
+   end-to-end against an Easy 2000 TI. Shutdown-related instant commands are
+   opt-in via the `allow_shutdown` flag because the firmware does not
+   auto-restart on mains return. [Initial PR #3447]
+   * Put the CDC-ACM port into raw mode at startup. A freshly enumerated
+     port defaults to canonical mode, where the kernel buffers the firmware's
+     binary replies waiting for a newline that never arrives, so the first
+     poll timed out as "no reply to initial status poll" unless the port had
+     been set raw externally. [issue #3500]
+   * Retry the initial status poll a few times before bailing out, so the
+     driver starts cleanly under the stock systemd units when the device has
+     only just enumerated at boot, instead of leaning on a fatal exit and a
+     `Restart=on-failure` cycle. [issue #3500]
+   * Drain the firmware's handshake reply before the first status poll.
+     Some models (e.g. the Easy Pro 3200 VA GT) are half-duplex and dropped
+     the read command that was sent while the handshake reply was still in
+     flight. [issue #3500]
+   * Compute `battery.charge` as `raw / 2.55` so a fully charged battery
+     reports exactly 100 % instead of 100.2 %.
+   * Added a `va` option to override the apparent-power rating when the
+     firmware model id is shared between product lines, correcting
+     `ups.power.nominal`, `ups.realpower.nominal` and the computed
+     `ups.load`. [issue #3500]
+
  - `richcomm_usb` driver updates:
     * Completed support for NUT standard USB configuration options, and fixed
       some problems possible when iterating devices. [issue #1768, PR #3477]
@@ -386,37 +415,6 @@ but the `nutshutdown` script would bail out quickly and quietly. [PR #3008]
      subdriver definition then). For consistency, these pointers are now
      assigned. [#1962]
 
- - Introduced a new NUT driver named `ragtech` which provides support for the
-   Ragtech "Easy Pro" family of line-interactive UPS units (also sold under
-   the NEP, TORO, INNERGIE and OneUP brands in Brazil). Devices present a USB
-   CDC-ACM serial interface (VID `0x04D8`, PID `0x000A`) and speak a proprietary
-   6-byte register-access protocol with read/write/AND/OR opcodes; the driver
-   covers the 20 models in the family-10 device table and was validated
-   end-to-end against an Easy 2000 TI. Shutdown-related instant commands are
-   opt-in via the `allow_shutdown` flag because the firmware does not
-   auto-restart on mains return. [PR #3447]
-
- - `ragtech` driver updates:
-   * Put the CDC-ACM port into raw mode at startup. A freshly enumerated
-     port defaults to canonical mode, where the kernel buffers the firmware's
-     binary replies waiting for a newline that never arrives, so the first
-     poll timed out as "no reply to initial status poll" unless the port had
-     been set raw externally. [issue #3500]
-   * Retry the initial status poll a few times before bailing out, so the
-     driver starts cleanly under the stock systemd units when the device has
-     only just enumerated at boot, instead of leaning on a fatal exit and a
-     `Restart=on-failure` cycle. [issue #3500]
-   * Drain the firmware's handshake reply before the first status poll.
-     Some models (e.g. the Easy Pro 3200 VA GT) are half-duplex and dropped
-     the read command that was sent while the handshake reply was still in
-     flight. [issue #3500]
-   * Compute `battery.charge` as `raw / 2.55` so a fully charged battery
-     reports exactly 100 % instead of 100.2 %.
-   * Added a `va` option to override the apparent-power rating when the
-     firmware model id is shared between product lines, correcting
-     `ups.power.nominal`, `ups.realpower.nominal` and the computed
-     `ups.load`. [issue #3500]
-
  - `riello_ser` and `riello_usb` driver updates:
    * Since the beginning, these drivers fenced availability of *either*
      `load.*` *or* `shutdown.return` instant commands based on current