Cypress: us_ticker fixes

[kyle-cypress]

Summary of changes

• Explicitly disable the timer used for the us_ticker before attempting to enter DeepSleep.
  A running timer will block DeepSleep, which would normally be good because we don't want the timer to accidentally lose counts. We don't care about that for us_ticker (If we're requesting deepsleep the upper layers already determined that they are okay with that), so explicitly stop the us_ticker timer before we go to sleep and start it back up afterwards.
• Add optional post-bsp-init hook
  This allows the application to inject its own resource reservations immmediately after the BSP (and therefore HAL) is initialized, ensuring that they can claim require resources before mbed tries to use them for more flexible purposes. For example, the application might want to claim a particular timer to make sure that it doesn't get picked for us_ticker (which can use any arbitrary timer instance).

Impact of changes

NA

Migration actions required

NA

Documentation

NA

---

Pull request type

[x] Patch update (Bug fix / Target update / Docs update / Test update / Refactor)
[] Feature update (New feature / Functionality change / New API)
[] Major update (Breaking change E.g. Return code change / API behaviour change)

---

Test results

[] No Tests required for this change (E.g docs only update)
[] Covered by existing mbed-os tests (Greentea or Unittest)
[x] Tests / results supplied as part of this PR

All failures are either transient or pre-existing
CY8CPROTO_062S3_4343W_GCC_ARM-NET.txt
CY8CPROTO_062S3_4343W-GCC_ARM.txt
CY8CKIT_062_WIFI_BT-GCC_ARM.txt
CY8CKIT_062_WIFI_BT-GCC_ARM-NETWORK.txt
CY8CPROTO_062_4343W_GCC_ARM-NET.txt
CY8CPROTO_062_4343W-GCC_ARM.txt

---

Reviewers

@ARMmbed/team-cypress

---

[ciarmcom]

@kyle-cypress, thank you for your changes.
@maclobdell @ARMmbed/mbed-os-maintainers please review.

[kjbracey]

Makes sense to me. I like the justification for "post-bsp init" hook better than the equivalent hook when it was put in for STM - there was a vague idea about it being for "extra RAM init" then, which wouldn't really work.

[0xc0170]

Makes sense to me. I like the justification for "post-bsp init" hook better than the equivalent hook when it was put in for STM - there was a vague idea about it being for "extra RAM init" then, which wouldn't really work.

cc FYI @ARMmbed/team-st-mcd

[0xc0170]

CI started

[mbed-ci]

Test run: SUCCESS

Summary: 6 of 6 test jobs passed
Build number : 1
Build artifacts