Errors: shake up how error digests are triggered, catch ints supplied as pin labels.

[Gadgetoid]

This change is mostly to shake up how errors are presented when calling gpiodevice.find_chip_by_pins using outdated, integer pin numbers. IE:

gpiodevice.find_chip_by_pins(1) will produce the errors:

Woah there, suitable gpiochip not found!
  ⚠️   /dev/gpiochip4: 1 is an int and has been skipped, did you mean "PIN1" or "GPIO1"?
  ⚠️   /dev/gpiochip3: 1 is an int and has been skipped, did you mean "PIN1" or "GPIO1"?
  ⚠️   /dev/gpiochip2: 1 is an int and has been skipped, did you mean "PIN1" or "GPIO1"?
  ⚠️   /dev/gpiochip1: 1 is an int and has been skipped, did you mean "PIN1" or "GPIO1"?
  ⚠️   /dev/gpiochip0: 1 is an int and has been skipped, did you mean "PIN1" or "GPIO1"?

[github-actions]

Pull Request Test Coverage Report for Build 6897064722

• 25 of 33 (75.76%) changed or added relevant lines in 4 files are covered.
• 4 unchanged lines in 1 file lost coverage.
• Overall coverage increased (+7.0%) to 41.86%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
tests/conftest.py	5	7	71.43%
gpiodevice/init.py	6	12	50.0%

Files with Coverage Reduction	New Missed Lines	%
gpiodevice/init.py	4	25.74%

Totals
Change from base Build 6731862195:	7.0%
Covered Lines:	90
Relevant Lines:	215

---

💛 - Coveralls