Fix: target discovery logic in stm32l4

[ALTracer]

Detailed description

• This is a couple bugfixes.
• The existing problem exhibits as invalid CCM RAM size on STM32G474CE.
• The PR solves it by changing the target->priv field behind typecasts to target->target_storage correct one.

Dereferencing that priv and typecasting it to stm32l4_priv_s results in unpredictable CCM sizes in BMDA and large sizes in BMF. It's for cortexm_priv_s, and the rest of this driver uses target_storage, so fix the last use of it.
Other dereferences of both priv and target_storage in other TUs seem to be okay per Qt Creator lookups (check them!)

Also, *device is not going to be NULL, it can be the 18th sentinel entry, but its device_id can be zero -- so correct the no-match branch condition which should trigger a second LUT walk.

Tested on gd32f103cb bluepillplus as bluepill.ini/stlink platform and stm32g474ceu target, BMDA as well. Gentoo gcc-13, crossdev. In particular, RTT autosearch no longer hangs up faulting in nonexistent CCM and retrying for ages.

Your checklist for this pull request

• I've read the Code of Conduct
• I've read the guidelines for contributing to this repository
• It builds for hardware native (see Building the firmware)
• It builds as BMDA (see Building the BMDA)
• I've tested it to the best of my ability
• My commit messages provide a useful short description of what the commits do

Closing issues

[dragonmux]

LGTM, merging. Thank you for the contribution!