fix(input.temp): Ignore EAGAIN on unavailable thermal-zones

[01110111000001]

Summary

When the hardware device is sleeping, the temp input plugin completely hangs waiting for the os.ReadFile call to return.
The EAGAIN returned by the kernel is handled internally by os.ReadFile and the function waits for the file to become available.
-> Open the file in non-blocking mode and silently ignoring unavailable thermal-zones.

Some comments/potential issues

1. It seems to me the EAGAIN could also happen on hwmon files but I do not have a device that exhibits this behavior. Should I call the new readFileAsync function for the hwmons too?
2. I fear some metrics from zones could be ignored when an EAGAIN is returned for another reason (like the kernel being busy but the hardware being available, if it could even happen?)
3. the silent dropping causes the number of metrics gathered to not be predictible as some thermal-zones might be present but not available and thus dropped. Should a default value be returned instead of dropping the metrics ? If so, what should be the default value? User defined in the config?

Looking forward to your feedback and your thoughts on the potential issues.
Thanks for your great work on this tool!

Checklist

• No AI generated code was used in this PR
• AI generated code used in this PR follows the InfluxData Policy on AI-Generated Code Contributions

AI was used to generate a first version (heavily modified since) of the newly added unit test for the readFileAsync function.

Related issues

resolves #18899

[01110111000001]

The integration tests failed at plugins/outputs/graphite/graphite_test.go:636, not sure why as I did not touch that file and the test does not seem to rely on the inputs.temp plugin.
I may be wrong though.

Tell me if should do anything to correct that!

[srebhan]

Thanks @01110111000001! Just one question about the content length...

[telegraf-tiger]

Download PR build artifacts for linux_amd64.tar.gz, darwin_arm64.tar.gz, and windows_amd64.zip.
Downloads for additional architectures and packages are available below.

☺️ This pull request doesn't significantly change the Telegraf binary size (less than 1%)

📦 Click here to get additional PR build artifacts

Artifact URLs

. DEB	. RPM	. TAR . GZ	. ZIP
amd64.deb	aarch64.rpm	darwin_amd64.tar.gz	windows_amd64.zip
arm64.deb	armel.rpm	darwin_arm64.tar.gz	windows_arm64.zip
armel.deb	armv6hl.rpm	freebsd_amd64.tar.gz	windows_i386.zip
armhf.deb	i386.rpm	freebsd_armv7.tar.gz
i386.deb	ppc64le.rpm	freebsd_i386.tar.gz
mips.deb	riscv64.rpm	linux_amd64.tar.gz
mipsel.deb	s390x.rpm	linux_arm64.tar.gz
ppc64el.deb	x86_64.rpm	linux_armel.tar.gz
riscv64.deb		linux_armhf.tar.gz
s390x.deb		linux_i386.tar.gz
		linux_mips.tar.gz
		linux_mipsel.tar.gz
		linux_ppc64le.tar.gz
		linux_riscv64.tar.gz
		linux_s390x.tar.gz

[01110111000001]

@srebhan I rebased my branch on top of the new master branch state so the comparison shows many changes.

I modified the comment to be more descriptive as well as the test case to clearly show only the first 8 bytes are read.

Thanks for your reviews!