docs(changelog): merge startup config entries (#11376)

The hole-punching and daemon-config highlights describe the same
user-facing change, a named startup error instead of an abrupt exit,
so fold them into one entry and drop the internal `log.Fatal` wording.

Author: lidel

docs/changelogs/v0.43.md

@@ -11,8 +11,7 @@ This release was brought to you by the [Shipyard](https://ipshipyard.com/) team.
 - [Overview](#overview)
 - [🔦 Highlights](#-highlights)
   - [🛜 One-time notice when behind CGNAT](#-one-time-notice-when-behind-cgnat)
-  - [🕳️ Clearer error when hole punching is misconfigured](#-clearer-error-when-hole-punching-is-misconfigured)
-  - [Daemon configuration errors no longer exit abruptly](#daemon-configuration-errors-no-longer-exit-abruptly)
+  - [🛑 Clearer errors for invalid config at startup](#-clearer-errors-for-invalid-config-at-startup)
   - [🔑 `ipfs config replace` keeps PeerID and private key in sync](#-ipfs-config-replace-keeps-peerid-and-private-key-in-sync)
   - [📦️ Dependency updates](#-dependency-updates)
 - [📝 Changelog](#-changelog)
@@ -40,13 +39,11 @@ Silence the notice with [`Internal.CGNATCheck`](https://github.com/ipfs/kubo/blo
 > [!IMPORTANT]
 > The daemon now refuses to start when [`Ipns.RecordLifetime`](https://github.com/ipfs/kubo/blob/master/docs/config.md#ipnsrecordlifetime) is shorter than [`Ipns.RepublishPeriod`](https://github.com/ipfs/kubo/blob/master/docs/config.md#ipnsrepublishperiod): records would expire before the republisher refreshes them, leaving the name unresolvable. Raise `Ipns.RecordLifetime` or lower `Ipns.RepublishPeriod` so the lifetime is at least the period.
 
-#### 🕳️ Clearer error when hole punching is misconfigured
+#### 🛑 Clearer errors for invalid config at startup
 
-Setting [`Swarm.EnableHolePunching`](https://github.com/ipfs/kubo/blob/master/docs/config.md#swarmenableholepunching) to `true` while [`Swarm.RelayClient.Enabled`](https://github.com/ipfs/kubo/blob/master/docs/config.md#swarmrelayclientenabled) is `false` is an unsupported combination: hole punching needs the relay client to coordinate the upgrade from a relayed to a direct connection. Kubo now refuses to start with an error that names both settings, instead of exiting abruptly.
+When startup config is invalid, Kubo now stops with an error that names the problem instead of exiting abruptly. One case is an unsupported hole-punching setup: [`Swarm.EnableHolePunching`](https://github.com/ipfs/kubo/blob/master/docs/config.md#swarmenableholepunching) set to `true` while [`Swarm.RelayClient.Enabled`](https://github.com/ipfs/kubo/blob/master/docs/config.md#swarmrelayclientenabled) is `false`. Hole punching needs the relay client to coordinate the upgrade from a relayed to a direct connection, so the error now names both settings.
 
-#### Daemon configuration errors no longer exit abruptly
-
-Invalid daemon configuration now returns startup errors through the normal command path instead of calling `log.Fatal`. This covers deprecated provider settings, removed providing options, delegated routing with providing enabled, private-network auto routing, and the removed `IPFS_REUSEPORT` environment variable.
+The same path also covers deprecated `Provider` and `Reprovider` settings, removed providing options, delegated routing with providing enabled, private-network `auto` routing, and the removed `IPFS_REUSEPORT` environment variable.
 
 #### 🔑 `ipfs config replace` keeps PeerID and private key in sync