vulkan: Set k_load_shmem to false when K is too large

[jeffbolznv]

See https://github.com/ggml-org/llama.cpp/pull/19075/changes#r2726146004.

Z:\github\jeffbolznv\llama.cpp\build\bin\RelWithDebInfo>llama-bench.exe -fa 1 -m c:\models\GLM-4.7-Flash-Q4_K_M.gguf -p 512 -n 128 -d 0,4096,16384
ggml_vulkan: Found 1 Vulkan devices:
ggml_vulkan: 0 = NVIDIA GeForce RTX 5090 (NVIDIA) | uma: 0 | fp16: 1 | bf16: 1 | warp size: 32 | shared memory: 49152 | int dot: 1 | matrix cores: KHR_coopmat
| model                          |       size |     params | backend    | ngl | fa |            test |                  t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| deepseek2 30B.A3B Q4_K - Medium |  16.88 GiB |    29.94 B | Vulkan     |  99 |  1 |           pp512 |      4434.49 ± 44.89 |
| deepseek2 30B.A3B Q4_K - Medium |  16.88 GiB |    29.94 B | Vulkan     |  99 |  1 |           tg128 |        231.50 ± 0.39 |
| deepseek2 30B.A3B Q4_K - Medium |  16.88 GiB |    29.94 B | Vulkan     |  99 |  1 |   pp512 @ d4096 |        907.02 ± 0.94 |
| deepseek2 30B.A3B Q4_K - Medium |  16.88 GiB |    29.94 B | Vulkan     |  99 |  1 |   tg128 @ d4096 |        192.08 ± 0.64 |
| deepseek2 30B.A3B Q4_K - Medium |  16.88 GiB |    29.94 B | Vulkan     |  99 |  1 |  pp512 @ d16384 |        261.88 ± 0.30 |
| deepseek2 30B.A3B Q4_K - Medium |  16.88 GiB |    29.94 B | Vulkan     |  99 |  1 |  tg128 @ d16384 |        118.39 ± 0.21 |

k_load_shmem after

Z:\github\jeffbolznv\llama.cpp\build\bin\RelWithDebInfo>llama-bench.exe -fa 1 -m c:\models\GLM-4.7-Flash-Q4_K_M.gguf -p 512 -n 128 -d 0,4096,16384
ggml_vulkan: Found 1 Vulkan devices:
ggml_vulkan: 0 = NVIDIA GeForce RTX 5090 (NVIDIA) | uma: 0 | fp16: 1 | bf16: 1 | warp size: 32 | shared memory: 49152 | int dot: 1 | matrix cores: KHR_coopmat
| model                          |       size |     params | backend    | ngl | fa |            test |                  t/s |
| ------------------------------ | ---------: | ---------: | ---------- | --: | -: | --------------: | -------------------: |
| deepseek2 30B.A3B Q4_K - Medium |  16.88 GiB |    29.94 B | Vulkan     |  99 |  1 |           pp512 |      5645.01 ± 46.79 |
| deepseek2 30B.A3B Q4_K - Medium |  16.88 GiB |    29.94 B | Vulkan     |  99 |  1 |           tg128 |        230.90 ± 0.37 |
| deepseek2 30B.A3B Q4_K - Medium |  16.88 GiB |    29.94 B | Vulkan     |  99 |  1 |   pp512 @ d4096 |      2941.50 ± 12.58 |
| deepseek2 30B.A3B Q4_K - Medium |  16.88 GiB |    29.94 B | Vulkan     |  99 |  1 |   tg128 @ d4096 |        192.27 ± 0.35 |
| deepseek2 30B.A3B Q4_K - Medium |  16.88 GiB |    29.94 B | Vulkan     |  99 |  1 |  pp512 @ d16384 |       1206.25 ± 2.11 |
| deepseek2 30B.A3B Q4_K - Medium |  16.88 GiB |    29.94 B | Vulkan     |  99 |  1 |  tg128 @ d16384 |        118.67 ± 0.27 |