drop 4 patches based on RC review feedback

drop 4 patches based on RC review feedback

Dropped patches:

- "bpf, riscv: Remove redundant bpf_flush_icache() after pack allocator finalize"
  Queues: 6.12, 6.6
  Reason: build break on riscv — removes the inline bpf_flush_icache()
  definition but the prerequisite (6798668ab195, "riscv, bpf: Remove
  duplicated bpf_flush_icache()") is missing in 6.12/6.6, leaving an
  undefined caller in arch/riscv/net/bpf_jit_comp64.c.
  Report: https://lore.kernel.org/stable/b8ef921f-e2ab-4cfb-b75c-89e4277214cf@oracle.com/

- "fuse: new work queue to periodically invalidate expired dentries"
  Queues: 6.18
  Reason: FUSE maintainer Miklos Szeredi NAK — "This is not stable
  material, and I don't think the dependency is real."
  Report: https://lore.kernel.org/stable/CAOssrKcqmVW3kJ131tRF3LCJVQUtdRB31B5HENDfpgrq6r=jEA@mail.gmail.com/

- "dcache: export shrink_dentry_list() and add new helper d_dispose_if_unused()"  [dependency]
  Queues: 6.18
  Reason: only present to support the dropped workqueue patch (Stable-dep-of: 5a6baf204610).

- "fuse: fix uninit-value in fuse_dentry_revalidate()"  [dependency]
  Queues: 6.18
  Reason: Stable-dep-of relationship to dropped workqueue patch; per
  maintainer the dependency isn't real and a trivial conflict resolution
  is the correct path — fix will be reposted directly to stable.

- "fuse: fix race when disposing stale dentries"  [dependency]
  Queues: 6.18
  Reason: Fixes: ab84ad597386 — cannot apply without the dropped
  workqueue patch (fuse_dentry_tree_work doesn't exist).

- "fuse: make sure dentry is evicted if stale"  [dependency]
  Queues: 6.18
  Reason: Fixes: ab84ad597386 — modifies fuse_dentry_tree_work()
  introduced by the dropped workqueue patch.

- "i3c: master: Fix error codes at send_ccc_cmd"
  Queues: 7.0, 6.18, 6.12, 6.6, 6.1, 5.15, 5.10
  Reason: prerequisite patches called out in the commit message
  ("Move rstdaa/entdaa/bus_init error suppression") are not all
  backported, so the semantics of the change differ from upstream.
  Report: https://lore.kernel.org/stable/a6d725bd-9ed4-4328-b2ba-078f658fd8b2@oracle.com/

- "fuse: fix premature writetrhough request for large folio"
  Queues: 7.0, 6.18
  Reason: Reviewed-by Joanne Koong (author of the parent commit that
  introduced large-folio writethrough support) reports the bug is not
  triggerable as fuse hasn't enabled large folios yet.
  Report: https://lore.kernel.org/stable/CAJnrk1bWJfT5hswsJ3X5HMZ=XLz5MO3pYXg6WkuYYvZ6ccS0Nw@mail.gmail.com/

Signed-off-by: Sasha Levin <sashal@kernel.org>